SCROLL NEWS / DISCOVERY
Search headlines
Find the stories shaping the conversation.
Results for " token " 100 found 🔀 AI-powered Shuffle
Sign in to save shuffle history.
Shaky political science: using "simulated" elections over real-world elections
[Dear DSOS readers – this article is the latest in our “Shaky Political Science” series. In our 38 page report, “Shaky political ‘science’ misses mark on ranked choice voting,” released in early December, we examined over 40 studies focused on ranked choice voting. We found that many of these studies suffered from puzzling research methodologies, poorly constructed surveys, unrealistic simulated elections, cherry picked data, and faulty analyses that often were contradicted by results from real-world elections (a link to our full paper is here). Here are links to the first three articles in our series: the first one summarizing the overall results of our report, the second one analyzing a study co-authored by University of Minnesota academic Larry Jacobs, which was one of the most error-prone of all the research we reviewed in our report, and a third article showing an admirable example of two political scientists holding two other political scientists accountable for their flawed, sloppy research]. In this fourth article in our “Shaky Political Science” series, we shine a spotlight on seven studies that were web-hosted and funded by New America’s Political Reform Program and its Electoral Reform Research Group (the ERRG is an ad hoc committee headed by New America’s Political Reform Program and joined by the American Enterprise Institute, the Unite America Institute and Stanford University’s Center on Democracy, Development, and the Rule of Law). This research consortium has been led by Lee Drutman, senior fellow with New America’s Political Reform Program. These seven studies are simply not credible from any objective standard about how political “science” is supposed to illuminate our understanding of either ranked choice voting or politics in general. The feature that six of these seven studies have in common is that they all used surveys – artificial, simulated elections -- to attempt to say something cogent or interesting about ranked choice voting (RCV) elections. None of them used real world data from actual RCV elections – either the 1000+ elections in the US or the many thousands of elections from places like Australia, Ireland and elsewhere where RCV has been used for 100 years. While a few of these studies conclude with some mixed positives about RCV, our analysis here is content neutral. The idea that, with a political reform like RCV, you can ask respondents in a mock, artificial election to vote with that method, with only the barest of explanation about how RCV works or its likely impacts (i.e. no spoiler candidates, majority winners, voters liberated to pick their favorite candidates, coalition-building), and then ask them their thoughts/reactions about voting in that system, with fake candidates and no real campaigning, no real exposure to an actual election or actual candidates, rests on some very puzzling assumptions. The findings of these seven studies actually are contradicted by the findings from many other studies based on real-world RCV elections. These New America studies also failed to list the real-world RCV studies in their own citations, instead including those studies that reinforced their conclusions. The seven studies featured in this article were chosen because they form a cohesive cohort which precisely illustrates one of the main points of our research paper, namely that using surveys, i.e. artificial elections, or mathematical/theoretical models, instead of real-world data, is in fact “shaky political science.” Below are summaries of our analysis of five of the seven New America studies. You can read our complete analysis of all seven studies at our paper, “Shaky political ‘science’ misses mark on ranked choice voting.” Choosing to “Vote As Usual” by Andre Blais, Carolina Plescia, and Semra Sevi. New America. (Blais et al., 2021). This study, web-hosted and funded by New America’s Political Reform Program and its Electoral Reform Research Group, used two surveys in which respondents completed ballots for simulated elections using plurality, approval, RCV and point (Score) voting. In both cases the authors provided the respondents with only minimal explanation of each method. For example, the entire explanation for RCV was “For this vote, rank the candidates from your first to your last choice (you do not need to rank them all). This is called the ranked vote.” That’s it. The study did not provide any context or explanation to survey respondents about why she/he might want to rank multiple candidates. Participants were not even told that ballots would be tabulated using an “instant” runoff, not to mention any description of RCVs ability to eliminate the spoiler effect or liberate voters to rank their favorite candidates without fear of helping the “greater evil” candidate to win. That’s like taking an iPhone back in time to 1970 and asking someone if they like it better than their rotary phone without explaining how it works. Their first survey targeted states that held 2020 Super Tuesday Democratic primaries and used the real candidates’ names in the survey; the second survey used fictitious candidates. In other words, they surveyed respondents who had only ever participated in plurality elections, had no history or background with using RCV or any of the other methods, had no clue about its likely impacts, and did not communicate how the alternative methods would be counted, and then asked them: “How satisfied are you with using each system?” Predictably, the winner was plurality elections, the system used for president, Congress, gubernatorial and state legislative elections – in other words, the method they have been using, year after year, all their voting lives. With so little explanation of RCV and its impacts, this study is simply not credible in discovering anything useful about real-world RCV elections. It merely shows that a basic voter education campaign is needed when adopting a new election system. Additionally, the result contradicts many other post-election surveys following real RCV elections, asking opinions from voters who had just participated in a RCV election, that showed a great deal of satisfaction with RCV (see details on that in Studies 22-25 in our paper, as well as the four studies cited in the review for Study 28). Share Other studies on “satisfaction” and “understanding” make similar mistakes, in terms of basing their surveys on inadequate explanation to respondents and lack of education about context. How could any survey ever hope to measure accurately, in essence simulate a real-world election, without actual campaigning candidates, the media writing about those candidates, organizations endorsing candidates and voters being inundated with crucial information? More than anything, what these “studies” based on surveys show is that if you ask voters about something they don’t know very much about, and fail to adequately educate them, their responses will unsurprisingly reflect their confusion. Equally puzzling is that New America’s Electoral Reform Research Group, headed by political scientist Lee Drutman, found this to be a credible enough study that they not only published it but highlighted it in Drutman’s own discussion of research related to ranked choice voting. This pattern was repeated in study after study featured by New America’s Electoral Reform Research Group. For example in Ranked-Choice Voting and Political Expression: Voter Guides Narrow the Gap between Informed and Uninformed Citizens by Cheryl Boudreau, Jonathan Colner, and Scott MacKenzie, the researchers based their analysis on a survey which found that respondents who were asked to vote in a simulated RCV election, with minimal explanation of RCV or how it works, displayed less evidence of being informed about how to use RCV effectively, with the least informed voters ranking fewer candidates. As pointed out under the study above, in a real election voters are informed by a wide variety of methods, including public education by the election administration, by interested political groups, the media, and the candidates themselves who benefit from voters being aware of how to vote in the new system. In Support for Ranked-Choice Voting across Race and Partisanship by Joseph Anthony, David C. Kimball, Jack Santucci and Jamil Scott, the researchers used respondents from different parties and races voting in simulated plurality and RCV elections, who indicated their satisfaction with each voting method. Instead of using real world election results, the authors fell back on the easy, armchair and less reliable academic research methodology of making up mock elections. As with the other studies, little explanation was given of how RCV works or what its implications might be (electing majority winners or eliminating spoilers, for example). Unsurprisingly, this study found that respondents were not very satisfied with RCV. However, when the authors added an explanation that RCV has a tendency to elect more women and people of color, they found that this greater explanation increased the preference for RCV. The more voters were familiar with it, and the more they understood potential positive impacts – surprise! -- the more they were open to switching away from the status quo. Of note, the results from this study are thoroughly contradicted by a number of other studies whose data was obtained from real RCV elections with real voters. Those studies show a great deal of satisfaction with RCV. More “fake election” research An interesting wrinkle on this “research by fake elections” approach, which only served to illustrate how unreliable such “research” can become, was taken in Ranked-Choice Voting, Runoff, and Democracy: Insights from Maine and Other U.S. States by Joseph Cerrone and Cynthia McClintock_,_ also web-hosted and funded by New America’s ERRG. It gets high marks for creativity, but an eye-roll for thinking this could possibly add anything meaningful to our understanding of real world RCV elections. This study tried to determine whether RCV, plurality or two-round runoff elections provided greater voter satisfaction and acceptance of election results. To determine that, the authors conducted a survey asking if voters were satisfied with a fictional newspaper report of an election, with three different versions given to respondents – plurality, runoff and RCV, with variations of Republican/Democrat winners and variations of “come-from-behind” winners,” i.e. candidates not initially in first place but who win the “runoff.” Again, a minimal explanation of RCV or how it works was given, and the survey showed that respondents had the greatest unfamiliarity with RCV than the other methods. The unsurprising result? Respondents were much more satisfied with “come-from-behind” victories in a traditional two-round runoff election than in RCV, even though RCV is also a runoff system but with a single round of counting. Interestingly, the dissatisfaction with RCV disappeared for those who were more familiar with RCV. In fact, among those respondents who were “very familiar” with RCV, they were equally satisfied with RCV, runoff and plurality. Surprise, surprise. Again, what this study revealed, more than anything, is not only the inadequacy of the explanation for RCV given by the authors of this study, but the failure of the research format itself, thinking that such an artificial mock election could somehow simulate what actually happens in the real world in a real RCV election. Those four studies reviewed above are fitting examples of inadequate and shaky political “science.” Other studies reviewed in our paper from the New America/Electoral Reform Research Group cohort which also used mock elections included: Ranked-Choice Voting is an Acquired Taste by Joseph Anthony and David C. Kimball. New America. (Anthony et al., 2021). Ranked-Choice Voting is No Refuge for Extreme Candidates by Melissa Baker. New America. (Baker, 2021). Also included in this New America cohort is Does Ranked-Choice Voting Reduce Racial Polarization? by Yuki Atsusaka and Theodore Landsman, which actually did use real-world election results data but applied an odd and unconventional methodology to the data which relied on shaky assumptions and non-real-world scenarios that were too divorced from reality. To determine levels of “racially polarized voting,” the researchers imagined “ethnic parties,” by which they mean political parties for each ethnic group, which was especially odd since the election data they were using came from nonpartisan RCV races. But then they made an even odder assumption, namely that since race and ethnic characteristics of studied voters were not available, the authors tried to infer that information based on who voters selected and the racial make-up of voters in each district. This amounted to little more than vague and ambiguous guesswork dressed up with a bunch of mathematical equations used to give it a veneer of scientific appearance. Indeed, it seemed to assume racial polarization in order to prove racial polarization. Based on this questionable methodology, the authors concluded that RCV doesn’t reduce racial polarization. But the premise of this study missed an obvious reality: since RCV incentivizes more candidates to run and can result in more diverse fields of candidates, one would expect more racially-polarized voting in voters’ top rankings, not less, because more voters would have someone on that ballot who represents their racial or ethnic identity and more freedom to vote for their real first choice. By encouraging more diverse candidates and eliminating the spoiler effect, RCV facilitates voting for a favorite candidate of your own race and then selecting a second choice with the best likelihood of winning. The real measure is what voters do with their additional rankings – do they consider other candidates and vote “for the best of the rest”? This study didn’t tell us anything useful about that. But readily available data shows this is common in real RCV elections, and candidates act on this knowledge by engaging with voters outside their racial group. So the authors of this study failed to consider or account for what actually happens in real-world elections, and their conclusions are tainted by this failure. New America has contributed a great deal over the past 2.5 decades to our understanding of many policy areas, including politics and political reform (full disclosure: Steven Hill, one of the authors of this article and of the “Shaky Political Science” research paper and series, is a former senior fellow and former director of New America’s political reform program, 2005-2010). So it was especially disappointing that New America funded and published such a large batch of dubious studies about ranked choice voting. The stated praiseworthy goal of its Electoral Reform Research Group was to inspire new academic research about ranked choice voting that would shed light and help the public better understand the impacts of RCV. Instead, the ERRG research did just the opposite. It produced dubious research that created more confusion and misinformation. In so doing, New America’s political reform program contributed to “shaky political science.” Steven Hill and Paul Haughey Share Upgrade to $5 subscription Thanks for reading DemocracySOS, a reader-supported digital portal for the pro-democracy movement. Subscribe for only $5 per month to receive full benefits and to support our work. Subscribe
AMD Tackles NVIDIA’s $4679 DGX Spark AI PC With Its $3999 Ryzen AI Halo: Now Available With 128 GB Memory For Blazing Fast LLMs
AMD has officially launched its Ryzen AI Halo AI PC, which offers super-fast token throughput at a $3999 price point.
Apple Stock Gains as Macs Attack the Token Meter
Local inference replaces recurring cloud charges with hardware whose utilization determines the savings.
I let Gemini in Google Maps plan my day and it went surprisingly well
Tired: tokens. Wired: tacos.
Are Tokens The New Currency? A Primer For Business
How CEOs can effectively harness token use for business success, while managing practices such as tokenmaxxing.
New CSS Attacks Can Break Webmail Defenses to Steal Passwords and Tokens
Outlook CSS techniques can spoof Microsoft sign-in and capture passwords, while Gmail image-set() can trigger external requests from malicious email c
Anthropic ships major Claude Design overhaul with design system imports, code round-trips, and a fix for its token-burning problem
Anthropic has overhauled Claude Design with brand-compliance controls, Claude Code integration, lower token usage and new enterprise app exports, positioning the AI tool as a serious platform for design-to-code workflows.
Trump receives inaugural 'America First award' from Republican Party
President Donald Trump received the first ever 'America First award' on March 25, presented by the Republican Party as a token of appreciation.
Anthropic doubles estimate for Claude Code token spend
Anthropic now says that the average cost per developer per active day is $13, up from $6 earlier this month.
(BPRW) NASA Postdoctoral Program Seeks Early Career and Senior Scientists for Prestigious Fellowships at Its Locations Across the U.S.
(Black PR Wire) OAK RIDGE, Tenn.--(BUSINESS WIRE)-- NASA invites both early-career and senior scientists to explore fellowship opportunities in its postdoctoral program and share in NASA's mission to develop advancements in science, te...
Tokenomics will lead to opportunities in 'edge AI,' Citrini says
Citrini Research predicts that the economics of tokenmaxxing will prompt a shift to "edge AI" that will have implications for investors.
Google's new Gemma 4 open AI model is sized for your laptop
Gemma 4 12B uses a new encoding scheme and token prediction to punch above its weight.
Google I/O kicks off with reveals on AI models, Gemini product expansion, and agentic features
Google I/O 2024: Alphabet highlights Gemini AI surge (3.2 quadrillion tokens, 900M MAUs), unveils Ask YouTube + Omni Flash video AI.
(BPRW) ProcuraFind® and Ally Financial Announce Strategic Initiative to Advance Small Business Growth Through Education, Readiness, and Opportunity
(Black PR Wire) Charlotte, N.C. — ProcuraFind® and Ally Financial announced a strategic, multi-year agreement focused on helping small businesses build stronger companies through education, financial empowerment, supplie...
Stable Sea Expands WisdomTree Relationship With Two New Tokenized Funds for Business Cash
/PRNewswire/ -- Stable Sea today announced it is expanding its strategic relationship with WisdomTree (NYSE: WT), a global asset manager with more than $150...
Baking soda & vinegar volcanoes with Mad Science of Austin & SA
30 Graduation Gift Ideas
Whether the grad you have in mind is completing 8th, 12th, or beyond….whether you want to buy a small token or something significant…you’ll find a graduation gift pick below. It’s often hard to think of a great graduation gift. Sometimes you just want to give a token to show that you’re proud, and sometimes graduation […]
Crypto stocks and tokens drop after Senate blocks landmark bill
Senators from both parties blocked a landmark digital asset market structure bill in a procedural vote Tuesday, sending cryptocurrency stocks tumbling after a major setback for industry. The long-stalled legislation
Runable hits $21M to bet AI agents can go from building businesses to growing them
Runable says 60%–70% of its 1 trillion-plus token usage in the last 90 days came from paying customers.
What causes sleep paralysis; health indicators of Alzheimer's diagnosis; sober October; plus other stories with Vanderbilt Health sources
(iStock) Food and Wine interviewed Peter Martin, MD, Professor of Psychiatry and Behavioral Sciences, for a story about sober October. William Schaffner, MD, Professor of Preventive Medicine, was quoted in CNN International’s Connect the World with Becky Anderson and NewsNation Morning in America for stories about the reported plague situation in Russia, after a laboratory worker at an institute in Siberia dedicated to studying the plague and other infectious diseases died of a mysterious illness. Prevention interviewed James McGuirk, MD, Assistant Professor, Division of Sleep Medicine, for a story about what causes sleep paralysis. Modern Healthcare interviewed Daniel Fabbri, PhD, Associate Professor of Biomedical Informatics and Computer Science, for a story about how health systems are managing AI token costs. WNKY interviewed Ziv Beckerman, MD, Associate Chief of Pediatric Cardiac Surgery, Surgical Director of Pediatric Heart Transplantation in the Golisano Pediatric Heart Institute at Monroe Carell Jr. Children’s Hospital at Vanderbilt about rapid recovery with extended ultra oxygenated preservation (REUP) that is a new form of preservation for heart organs. Ray Romano, PhD, MPH, MSN, Nurse Practitioner, Vanderbilt Memory & Alzheimer’s Center, was quoted Prevention about research that found other health issues may be an early indicator of Alzheimer’s.
The Quintessential Trump-era Politician
In 2020, I wrote an article about collaboration and collaborators. The title was History Will Judge the Complicit, and it started with the story of two famous East Germans, Markus Wolf and Wolfgang Leonhard, who both landed in East Berlin just after the end of the war. Although they came from very similar backgrounds, the former stuck with the regime, eventually becoming its spy boss; the latter, horrified by the brutal reality of Soviet-style communism, escaped. By way of juxtaposition, I also wrote about Lyndsey Graham and Mitt Romney. These two men also made very different decisions when confronted by Donald Trump’s assault on the ideals that they had both grown up believing: A glance at their biographies would not have led many to predict what happened next. On paper, Graham would have seemed, in 2016, like the man with deeper ties to the military, to the rule of law, and to an old-fashioned idea of American patriotism and American responsibility in the world. Romney, by contrast, with his shifts between the center and the right, with his multiple careers in business and politics, would have seemed less deeply attached to those same old-fashioned patriotic ideals. Most of us register soldiers as loyal patriots, and management consultants as self-interested. We assume people from small towns in South Carolina are more likely to resist political pressure than people who have lived in many places. Intuitively, we think that loyalty to a particular place implies loyalty to a set of values. But in this case the clichés were wrong. It was Graham who made excuses for Trump’s abuse of power. It was Graham—a JAG Corps lawyer—who downplayed the evidence that the president had attempted to manipulate foreign courts and blackmail a foreign leader into launching a phony investigation into a political rival. It was Graham who abandoned his own stated support for bipartisanship and instead pushed for a hyperpartisan Senate Judiciary Committee investigation into former Vice President Joe Biden’s son. It was Graham who played golf with Trump, who made excuses for him on television, who supported the president even as he slowly destroyed the American alliances—with Europeans, with the Kurds—that Graham had defended all his life. By contrast, it was Romney who, in February, became the only Republican senator to break ranks with his colleagues, voting to impeach the president. “Corrupting an election to keep oneself in office,” he said, is “perhaps the most abusive and destructive violation of one’s oath of office that I can imagine.” In the wake of Graham’s unexpected death on Saturday, I wrote about him again. He was, in many ways, the quintessential politician of Trump’s Washington, a man who shifted 180 degrees. Graham had been a fervent American patriot, proud reserve officer and military lawyer; he became one of the most important defenders of a president who believes soldiers are “suckers and losers” and shows no respect for law of any kind. As his politics changed, his language grew coarser and his behavior grew worse: He urged Israel to “level the place” in Gaza, and insulted the Danish prime minister in Munich, prompting a colleague, Senator Elissa Slotkin, to issue an apology. Here’s an excerpt from the article, which you can read in The Atlantic (gift link here) Like his close friend John McCain, Graham believed that America should stand at the center of a broad democratic alliance. The few times I met him were in that context, at the Munich Security Conference or at other European-American gatherings, which he once attended with great regularity. He also took seriously the practice of American democracy at home. In a 2014 conversation with The Atlantic, he described himself as a pragmatic politician who eschewed populist slogans. “I know Washington is broken, but what’s broken about it is everybody yelling and nobody trying to fix it,” he said. “I’m trying.” When Donald Trump first appeared in U.S. politics, Graham recognized him immediately for what he was: the spokesman for an alien ideology, one radically different from the idealistic patriotism that Graham had practiced and preached since childhood. Trump privately mocked the military that Graham loved as nothing but “suckers and losers.” He put falsehoods, cynicism, and personal greed at the center of his politics, while expressing disdain for transparency, accountability, and democracy itself. In 2015, Graham described Trump as a “race-baiting, xenophobic, religious bigot” who should “go to hell.” Also as a “nutjob.” When Trump won, Graham understood, as did so many others, that he would have to make some important choices. For a while, he went silent. In the spring of 2016, I saw him at one of those conferences in Europe. He seemed too depressed to speak. But then, like many other Republicans—and, more important, like many other people who have lived under political occupation or experienced radical regime change—he made the decision to abandon his previous ideals, to bury the patriotism that was once so important to him, and to become, instead, a loud, opportunistic collaborator. As I wrote in 2020, it’s very hard to look in to anyone’s heart and to understand their motivations from the outside. But it does seem that Graham had a deep need to be relevant, to be in the game, to be a player, at any cost. And that, rather than any particular achievement, is what he will be remembered for. Continuing to monitor conflicts of interest, ostentatious emoluments, outright corruption and policy changes that will facilitate outright corruption. (Read my original article, Kleptocracy Inc and check out the SNF Agora Institute chart) June 27 Trump diverted funds from more than 900 projects nationwide—including safety upgrades at national parks—to finance beautification projects around Washington, including $700,000 used to fix a White House walkway. June 28 After a years-long Department of Justice investigation into Abbott Laboratories over its management of a baby formula facility where potentially deadly bacteria turned up enough evidence to criminally charge the firm, top department officials instead opted for a financial penalty—illustrating the department’s move away from strict corporate enforcement under the Trump administration. June 29 Defense Secretary Pete Hegseth stacked the Defense Policy Board—which is charged with providing independent national security recommendations to the Office of the Secretary of Defense—with Trump allies, including tech billionaire Marc Andreessen, who has invested in defense contractors like OpenAI and SpaceX, after purging the panel last year. Trump purchased between $1 million and $5 million worth of stock in Axon—which makes about 90% of US tasers—two weeks before ICE solicited bids for a $220 million taser contract. The White House unveiled limited-edition US passports for America’s 250th birthday, featuring a portrait of Trump overlaid on the Declaration of Independence. June 30 Trump made around $1.2 billion from his various crypto companies last year, including more than $500 million from World Liberty Financial and $600 million from sales of his meme coin. Trump is considering issuing 250 pardons to commemorate America’s 250th birthday. These are expected to be largely doled out through the Office of the Pardon Attorney’s informal network of intermediaries. White-collar defense attorneys say they can typically win pardons for clients for $2 million. Last year, Trump administration officials awarded a no-bid contract worth up to $500 million to Clark Construction for the White House ballroom—negotiations in which the president was directly involved. July 1 The Amazon MGM Studios documentary about Melania Trump earned the First Lady more than $10 million. Trump made hundreds of individual stock purchases worth as much as $12.8 million the day before his announcement pausing “Liberation Day” tariffs for 90 days sent the S&P 500 soaring nearly 10%—one of the largest single-day gains ever. Mar-a-Lago and Trump National Doral saw record surges in revenue during Trump’s first year in office, earning $77.5 million and $121.9 million respectively, compared to average annual revenues of $25 million and $75 million during Trump’s first term. Entities across the Persian Gulf paid more than $300 million to Trump’s businesses last year, including $263 million from the sale of half his stake in World Liberty Financial to a firm backed by Sheikh Tahnoon bin Zayed Al Nahyan, a top UAE royal and brother of the country’s president. Kash Patel failed to properly disclose his stock purchase of a Department of Justice contractor, MicroStrategy, worth up to $250,000 last year, despite laws requiring disclosure within 45 days of the trade. July 2 Trump has been directing administration officials to steer more government contracts—so far worth more than $2.4 billion—to Jeff Bezos’s Blue Origin aerospace firm, after Amazon gave $1 million to Trump’s inaugural fund, donated to the White House ballroom project, and reached a $40 million deal to license a Melania Trump documentary. Alongside the billions he made from cryptocurrency last year, Trump earned millions of dollars from real estate licensing fees, including $21 million from organizations in the UAE and $9.2 million in Saudi Arabia, along with smaller sums from merchandise sales, including $4.7 million from Trump Watches and $35,920 from 45 Guitar, a guitar brand endorsed by the president. House Democrats accused consultants tied to the Trump administration of financial fraud, alleging they gave prospective donors to America250—the bipartisan committee created by Congress to celebrate the nation’s semiquincentennial—the banking and routing numbers for Freedom 250, the White House’s fund for events like the UFC cage fight. Democrats also allege that Freedom 250’s CEO traveled to the World Economic Forum in Davos to solicit donations from foreign government officials and business leaders. The Trump administration’s proposed rollback of gun regulations would lift the prohibition on mailing handguns to individuals, a significant win for GrabAGun—the direct-to-consumer firearm retailer on whose board Donald Trump Jr. sits and in which he holds a 1.1% ownership stake. GrabAGun currently must rely on a middleman to transfer firearms to customers. Trump’s financial disclosure revealed that he made more than 21,000 securities trades during his first year back in office, growing his investment account to at least $858 million across stakes in around 1,600 companies, compared to the 13 total trades Joe Biden made during his entire presidency. July 4 Nearly 1 million people who bought the president’s meme coin—Trump Coin—collectively lost $3.8 billion on the token, while he walked away with $636 million. July 6 Shares of Dell closed up 4.4% after Trump recommended people buy the firm’s computers during a speech announcing the launch of Trump investment accounts for children, boosting the more than $1 million worth of shares the president owns in the company. July 9 Federal Reserve Chair Kevin Warsh named crypto investor and Trump ally Marc Andreessen to a task force aimed at modernizing the Federal Reserve, including rethinking its approach to inflation—a key factor keeping interest rates elevated. Palm Beach International Airport has officially changed its name to Trump International Airport—a move that will require Florida to pay royalties to the president’s family for licensing his name. I was there a few months ago, during a trip to Barcelona. The sculptures were a revelation, I’d never seen so many together before. Each one has its own personality. Enjoy the good weather, if you are lucky enough to have it. History Will Judge the Complicit The Quintessential Politician Autocracy Inc Open Letters, from Anne Applebaum is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber. Subscribe
What's Going On With AMC Entertainment Stock Friday?
AMC Entertainment gains Friday amid CEO Adam Aron's dispute with Robinhood over stock tokens. Read key technicals and price targets.
Cheaper AI models may strain power grid, consulting firms say
In new reports published this week McKinsey and BCG reveal the impact of falling token prices and impact on the grid.
How to manage token costs while keeping agentic goals on track
Learn about AI tokenomics and how Dell AI Factory with NVIDIA helps reduce token costs by running AI agents locally.
NinjaTech AI Launches Enterprise AI Workforce Platform Using Open-Weight Models
Built to end runaway per-token AI costs, the platform deploys into the customer’s own cloud and data centers, so data never leaves their control PALO ALTO, Calif., Sept. 28, 2026 — NinjaTech AI today launched SuperNinja Enterprise, an AI workforce platform that puts AI employees to work around the clock inside the customer’s own cloud, on […]
Chamath Palihapitiya: Some CFOs are in for a tokenmaxxing shock
AI spending ballooned as companies pushed workers to use more tokens. Chamath Palihapitiya says CFOs may soon feel the hangover.
Hunter Biden launching a laptop-themed cryptocurrency
Hunter Biden's $Laptop meme coin launches on Base blockchain, with 20% of tokens going to those who lost money on Trump's $TRUMP crypto coin.
Enterprise AI spending grows, OpenAI leads, RBC reveals
RBC's latest CIO survey finds no AI token panic, no SaaSpocalypse, and surging enterprise AI spending, led by OpenAI.
Shaky political "science": breakdown of the peer review process
[This article is the latest in our “Shaky Political Science” series. In our 38 page report, “Shaky political ‘science’ misses mark on ranked choice voting,” released in December 2025, we examined over 40 studies focused on ranked choice voting (RCV). Many of these studies suffered from puzzling research methodologies, poorly constructed surveys, unrealistic simulated elections, cherry picked data and faulty analyses that often were contradicted by results from real-world elections (a link to our full paper is here). Here are links to the first four articles in our series: the first one summarizing the overall results of our report; the second one analyzed a study co-authored by University of Minnesota academic Larry Jacobs, which was one of the most error-prone of all the research we reviewed in our report; a third article showing an admirable example of two political scientists holding two other political scientists accountable for their flawed, sloppy research; and a fourth article that analyzed seven studies sponsored by New America’s political reform program that deployed flawed methodology based on “simulated” elections instead of real-world elections, which when combined with some basic misunderstandings of RCV, produced odd and misleading results]. In this fifth article in our “Shaky Political Science” series, we examine two flawed studies on the impact of RCV on voter turnout authored by San Francisco State University professor Jason McDaniel. McDaniel’s research in these two papers stands out for its puzzling methodology, inadequate and cherry-picked data, and faulty analyses that not only misunderstands ranked choice voting but reflects a basic misunderstanding of politics in general. The first of the McDaniel studies actually passed through a peer review process and then was subsequently published by the Journal of Urban Affairs, despite its obvious errors. It is perplexing that a jury of peer political scientists did not catch the errors and send it back to McDaniel for a rewrite or outright rejection. This calls into question the credibility of the peer review process itself. The peer review process is supposed to represent the intellectual pinnacle of the profession – a way for academics to certify the quality of each other’s work, and to act as a check against sloppy research. In the case of the McDaniel study, the peer review process failed. Yet his flawed study has been cited dozens of times by other political scientists in their own flawed works — an unvirtuous circle of academic research. Here is the first Jason McDaniel study that we analyzed: “Writing the Rules to Rank the Candidates: Examining the Impact of Instant-Runoff Voting on Racial Group Turnout in San Francisco Mayoral Elections,” by Jason McDaniel, 2016, Journal of Urban Affairs. This study covered San Francisco’s use of ranked choice voting (RCV) in mayoral elections, purporting to show that RCV reduces voter turnout overall, increases voter errors, and has a negative impact on “marginal populations,” i.e. racial minorities, mainly as the result of the alleged complexity of RCV asking voters to rank candidates and having to be familiar with multiple candidates and related dynamics. McDaniel reached these conclusions by examining five San Francisco mayoral elections taking place in odd years (i.e. where there are no federal or state races on the ballot) from 1995 to 2011, comparing three earlier non-RCV races (in 1995, 1999 and 2003) with the later two RCV races (in 2007 and 2011). Only two RCV races were studied, and as McDaniel acknowledged, only the contest in 2011 was “competitive” (that contest in 2007 was the first mayoral election using RCV, and incumbent mayor Gavin Newsom won easily in the first round of counting, earning 74% of first choices and far ahead of the second place candidate who had just 6.3% of first choices). Yet even the allegedly competitive contest in 2011was won by an incumbent with a landslide margin of nearly 20 points. No other elections were studied by McDaniel, even though San Francisco elects 17 other offices besides mayor with RCV, including the 11 seats on the Board of Supervisors (the name for the city council) in even years, and six other citywide offices in odd years. Previous academic research has found that competitive elections can stimulate greater voter involvement and higher turnout. For unexplained reasons, this study ignored at least 34 other RCV races between the years 2004 to 2011, many of which were very competitive and had high voter turnout. Instead, McDaniel’s analysis based its conclusions on data from only a single, marginally-competitive mayor’s race using RCV. These 2007 and 2011 mayoral elections were the only two RCV elections that McDaniel used as the sole basis for his small data set to determine the impact of RCV on voter turnout. Despite having no meaningful data to inform his study, McDaniel nonetheless concluded that RCV lowered turnout when compared to the previous three mayoral non-RCV elections using two-round runoffs. Then, using the same inadequate database of San Francisco elections, McDaniel also found that “Instant-runoff voting is associated with a significant decline in Black voter turnout and White voter turnout…On average, Black voter turnout decreases by 18 points and White voter turnout decreases by 16 points …” But how could he credibly arrive at that conclusion when his study was based on voter turnout in two RCV contests that were not even remotely competitive? Beyond that, McDaniel tacitly admits that the key determining factor driving Black voter turnout was not RCV vs non-RCV elections, but the presence of a viable Black candidate. McDaniel wrote, “Black voter turnout was about 17 percentage points higher for the two elections that featured an African American candidate compared to the three elections that did not” — but one of those three elections did not use RCV. The same dynamic was present for White voter turnout, in which higher turnout was not correlated with RCV vs non-RCV elections, but the presence of a viable White candidate. Beyond that, McDaniel’s voter turnout results, whether correlated with the race of the candidates or RCV vs non-RCV elections, showed no clear patterns. In his Figure 1, he shows low Asian voter turnout in the non-RCV elections in 1999 and 2003 when there was no viable Asian candidate running, but in the 2007 mayoral election using RCV, there is still no viable Asian candidate, yet Asian turnout increases dramatically. For Latinos, voter turnout is virtually the same in 1999 (no viable Latino candidate, non-RCV elections), 2003 (leading Latino candidate, non-RCV elections), 2007 (no viable Latino candidate, RCV elections) and 2011 (leading Latino candidate, RCV elections). Meanwhile White voter turnout increased dramatically from a low in the 2007 mayoral election, in which the leading candidate was a white incumbent, to the 2011 election in which there were no leading white candidates that finished any higher than sixth place with 5.6% of the vote. Clearly other factors were determining voter turnout than either the race of the candidates or RCV vs non-RCV elections. Despite these inconsistent findings, McDaniel boldly marches forth to render conclusions about the impact of RCV on different groups of racial voters, even though his conclusions are not remotely supported by his data. This looks like the worst kind of “data cherry picking” to cover up a puzzling conclusion based on a poor data set involving only two RCV elections, neither of which were remotely competitive. McDaniel’s cherry picking didn’t stop there. In calculating voter turnout for the three non-RCV mayoral elections in 1995, 1999 and 2003, he failed to mention that in each of these years there was a two-round runoff election, with a first round election in November in which no candidate garnered a winning majority of the vote, followed by the decisive runoff in December. So there were two non-RCV elections in each election year. But McDaniel only decided to include one of the two elections in his data set, apparently the December runoff in each year (which had higher turnout in two out of the three mayor’s races, and lower turnout in the third). So he ignored the November 1995 election, the November 1999 election when turnout was only 45 percent, and the November 2003 election when turnout was only 45.7 percent. This is flawed methodology based on blatant cherry picking, since McDaniel simply omitted three elections worth of data that did not fit within his frame. His study should never have made it through a peer review process, and it is an embarrassment to political science that not only was it approved for publication, but it has been uncritically cited by dozens of other political scientists. Share A “study” without wider context or credible data Lacking a data set that included even a single competitive RCV mayoral race, McDaniel declared that RCV lowered voter turnout in 2011, when turnout was 42.5 percent. But he also ignored -- or did not realize -- that voter turnout in mayoral elections during 2011 had sharply declined in all big cities, including far more steeply in nearby Los Angeles and other cities. In fact, San Francisco’s election in 2011 had the second highest turnout of any mayoral election in the nation’s 22 largest cities from 2008-2011. Was 2011 just a particularly low turnout year all across the country, perhaps because the country was still reeling from the aftermath of the housing market collapse in 2008-2010? McDaniel ignored this factor and did not search for alternative explanations beyond his flawed thesis. Furthermore, while McDaniel suggested that RCV leads to more “overvotes” – which is an invalidated ballot in which a voter, in either a non-RCV or RCV race, makes a mistake and selects two candidates for a single choice – he failed to address that in the June 2012 US Senate primary contest, which used the typical “one-choice” plurality method and did not use RCV, there was more than five times as many overvotes in San Francisco and Oakland than in the RCV mayoral elections in those cities. Additionally, turnout had been significantly higher and more representative in RCV elections for the San Francisco Board of Supervisors races -- that McDaniel didn’t even bother to study -- as compared to previous “delayed two-round runoffs” for those offices, often higher by as much as 40% in the RCV races. Perhaps the most surprising irony was that an earlier study co-authored by McDaniel (Overvoting and the Equality of Voice under Instant-Runoff Voting in San Francisco by Francis Neely and Jason McDaniel, 2015, California Journal of Politics and Policy) found that overvotes are often more common in lower-income precincts, but that the pattern of overvoting is similar in both RCV and non-RCV contests. Wrote Neely and McDaniel: “[Overvote] errors appear to be a function of complexity in general and not IRV per se…. the pattern of overvoting is similar in both IRV and non-IRV contests…Even among votes cast in a top-of-the-ticket race like U.S. Senate, overvoting occurs in an uneven fashion” (IRV, or ‘instant runoff voting,’ is another name for ranked choice voting in single-winner contests). So McDaniel not only contradicted his own conclusion in his earlier study with Neely, but the earlier study had been an important corrective for one of the most frequently made mistakes made by anti-RCV political scientists – a continual failure to provide crucial context by pairing a comparative analysis of real world RCV elections with an analysis of real world plurality or other electoral methods. Unfortunately McDaniel’s earlier study does not get cited much by other researchers, and McDaniel did not even cite it in his own flawed study analyzed above. Incredibly, McDaniel cherrypicked McDaniel, even as his flawed study has been cited dozens of times. Despite being a textbook example of poor political “science,” featuring both inadequate as well as cherrypicked data, the shelf life of McDaniel’s study has been extended endlessly by other political scientists and media outlets who have cited it uncritically. Equally troubling, the Journal of Urban Affairs put this paper through its peer review process and published it despite its numerous flaws. McDaniel gets it wrong, Part II Not content with producing one shoddy, confused paper, McDaniel produced another one a few years later, here is the link: Electoral Rules and Voter Turnout in Mayoral Elections: An Analysis of Ranked-Choice Voting by Jason McDaniel. San Francisco State Univ. (McDaniel, 2019). In this paper, McDaniel extended his previous flawed research focused on San Francisco, about voter turnout and voter impact, to multiple RCV cities through the 2018 elections. In his data set he included San Francisco, Minneapolis, St Paul, Oakland, Berkeley, San Leandro and Santa Fe. He then compared their turnout in mayoral elections to non-RCV cities, and to mayoral elections in those cities held before first RCV usage, and concluded “a significant decrease in voter turnout of approximately 3–5 percentage points in RCV cities after the implementation of RCV.” But he replicated many of the same mistakes as his first study presented above. While in this study McDaniel attempted to distinguish between various factors, such as odd year vs. even year elections, competitive races vs. noncompetitive, contests with open seats and other factors, the analysis is superficial and inadequate. He attempts to incorporate so many often-conflicting variables that his conclusions are once again unsupported by the actual data, and ultimately unconvincing (this second study never underwent a peer review process). For example, McDaniel declared that for RCV cities “in elections that are more competitive than average (margin of victory less than 20%), there is no significant RCV effect on voter turnout.” Yet McDaniel is a professor in a city, San Francisco, in which the 2018 special election using RCV to fill a mayoral vacancy was exceedingly close, decided by a margin of only one percentage point, and resulted in the second highest turnout in San Francisco history for a mayoral election. Illustrating the complexity involved in assessing the critical factors in turnout, that special election was held on the same election day as federal and state offices, which tends to boost turnout. Yet more voters participated in that RCV mayoral election than voted on the same day in the non-RCV primary for governor in which former San Francisco mayor Gavin Newsom was running. Even more puzzling, in this study McDaniel inaccurately cited his own previous study from 2015 with Neely. McDaniel wrote, “Neely and McDaniel (2015) examine individual ballot image files within a natural experiment, and find higher rates of ballot errors in RCV elections compared to non-RCV elections” (emphasis added). However, we quoted from that 2015 study several paragraphs above, and in actual fact Neely and McDaniel actually concluded the opposite -- that the pattern of overvoting – which is the most common form of ballot error – is similar in both RCV and non-RCV contests. McDaniel also ignored the fact that in Bay Area RCV elections held in even years – at the same time as federal and state elections, when turnout is highest -- the “undervote” (that is, the dropoff in ballots cast for governor or mayor) has declined, which contradicts his thesis that the act of ranking ballots in RCV elections is too complex and presents a barrier for certain demographics of voters. To his credit, McDaniel actually acknowledged another contradiction in his own paper, which found a negative RCV turnout effect in some odd-year elections but not others, admitting that “may present a challenge to the theory that increased complexity [from RCV] will have a marginal negative effect on voter turnout.” But he tried to wallpaper over that contradiction by blaming it on an unlikely source -- bad election administration. McDaniels wrote, “This would suggest that the negative impact of RCV on voter participation may be alleviated by high quality election administration,” without presenting any data or even anecdotes to support such an unfounded theory. While competent election administration of any election, whether RCV or non-RCV, is crucially important, there is no credible research that supports the far-fetched notion that the quality of election administration has much of an impact on voter turnout in US elections, whether using RCV or otherwise. Upgrade to a $5 subscription The unvirtuous circle of academic research It is puzzling how McDaniel’s first study could have survived the peer review process for the Journal of Urban Affairs. The mistakes in McDaniel’s first study above should have been obvious to anyone with a basic understanding of ranked choice voting as well as elections in general, and elections in San Francisco specifically. It undermines faith in the credibility of the peer review process to think that a jury of peer political scientists did not catch the egregious errors and reject McDaniel’s sloppy research and paper. And it remains a black mark on political science that this thoroughly discredited study has been cited dozens of times by other researchers. Indeed, in our research paper it became apparent that many political scientists cite each other’s flawed research to pad their list of citations and to legitimize their own flawed research, creating an unvirtuous circle of political science. More broadly, McDaniel’s voter turnout studies, like other studies evaluating RCV’s impact on turnout, lacked a recognition that no credible opinion or research can claim that RCV increases voter turnout merely because so many voters are excited to vote for their favorite candidates instead of “lesser evil” candidates without worrying about spoilers. Instead, with respect to turnout, the advantage of RCV has always been that, in most places where it has been used, it has eliminated a pre-November primary or post-November runoff, which usually saw about half the turnout of the November general election. This is especially true during even years when there are state and federal elections on the ballot which substantially drive turnout more than local contests. Any study that does not adequately account for this non-November primary/runoff vs November general election dynamic, or odd year vs even year election factors, is simply not credible. Both of McDaniel’s studies largely failed on those grounds. Steven Hill and Paul Haughey Share Upgrade to a $5 subscription Thanks for reading DemocracySOS…your digital portal for the pro-democracy movement. Subscribe for only $5 per month to receive full benefits and to support our work. Subscribe
It’s Nearly Impossible to Budget for AI Spending
Artificial-intelligence use is measured in tokens, but the models’ use of tokens at any given task can be unpredictable.
AI Tokenomics Explained: What Every Business Leader Needs To Know
AI tokens determine how generative AI systems process information, calculate usage and generate costs, making tokenomics essential for effective AI budgeting.
Synthesia HR chief: Tokenmaxxing isn't the goal. Getting results is.
Companies are pushing workers to use AI. Synthesia's Laura Gonzalez says measuring token usage can reward activity over outcomes.
'What a joke': Github Copilot's new token-based billing spurs consternation among devs
The golden age of Microsoft's Github Copilot appears to be at an end.
Watch Robinhood's Kerbrat on Business Strategy, Crypto
Johann Kerbrat, Senior Vice President and General Manager of Crypto and International at Robinhood, discusses the company's expansion plans and investment opportunities in the cryptocurrency market. He speaks with Haslinda Amin on the sidelines of the Token2049 conference in Singapore.
Is Autodesk (ADSK) Using Small-Business Flex Pricing To Quietly Redefine Its Competitive Moat?
Earlier this month, Autodesk Inc. launched “Autodesk for Small Business,” introducing a Small Business Hub, product refinements across AutoCAD, Revit, Fusion, and Flow Studio, plus more flexible Autodesk Flex token and pricing options in the US and UK. The initiative materially lowers onboarding complexity and upfront commitment for small firms, potentially broadening Autodesk’s customer base while giving these users a clearer feedback channel into future product development. Next, we’ll...
N0va Phishkit Targets US and EU Businesses: A New Challenge for Identity Security
N0va phishing abuses legitimate authentication flows to capture access and refresh tokens and establish SSO access to corporate resources.
Bitcoin treasury companies shed $80bn in value as business model unwinds
Crypto hoarders have been selling tokens and reverting to former activities
Ripple Plans a Credit Business Backed With XRP. Does Lending Finally Give the Token a Use That Moves the Price?
Ripple's new credit service locks XRP as collateral for payment loans, a model that works very differently from the pass-through payments that have left many holders waiting for a price catalyst. Whether this finally creates sustained demand hinges on details Ripple has yet to reveal.
Hyperliquid Is Building A $1B Business, Says Grayscale – But Traders Just Got Reminded Of The Risks Amid SK Hynix Turmoil
HYPE's circulating supply is estimated at 270-310 million tokens by 2027, said Grayscale, influenced by the unlocking of vested tokens at a rate of 550,000 per month.
Cognition's CEO says token leaderboards are 'directionally correct'
Cognition CEO Scott Wu said there are better ways to incentivize employees to use AI.
NYT Strands Answers Today: Hints & Clues For Sunday, October 4 (Grab Your Tokens)
Looking for help with today's NYT Strands puzzle? Here's an extra hint to help you uncover the right words, as well as all of today's answers and Spangram.
Fake IT Calls Target Executives in Microsoft 365 Data Theft and Extortion Attacks
PREY-0058 uses help desk vishing and AitM token theft to steal data from Microsoft 365 and other SaaS services for extortion.
Call of Duty: Modern Warfare 4 Is Finally Making Double XP Tokens Only Active During Matches
Call of Duty: Modern Warfare 4 is making a major quality of life change that fans have been wanting to see for years.
Margaritaville Ball Drop 2027 Gala Megawatt Entertainment
Welcome NYE 2027, view the ball drop at a tropical paradise gala at Margaritaville Times Square Hotel, a family-friendly event with VIP amenities
ConsentFix and ClickFix: How Microsoft 365 Accounts are Hijacked in 3 Seconds
ConsentFix and ClickFix attacks steal Microsoft 365 tokens in seconds using fake prompts and OAuth flows. Learn how these MFA bypass tactics work and how to defend against them.
How to scale AI without runaway inference costs
Learn how tokenomics, on-device AI, and smarter model orchestration can help enterprises scale AI sustainably.
Investors are actually underestimating the AI boom, Goldman Sachs says
The bank expects token consumption to increase 24 times by 2030, largely driven by the rise of enterprise agents.
Palantir Just Issued a Warning to Business Leaders About AI: Guard This 1 Thing Like 'Treasure'
Palantir also warned about tokenmaxxing and a sense of 'false progress.'
How A.I. Tokens Work
As large language models become more advanced, they use ever more resources. Who will foot the bill?
Perplexity launches a local AI agent with zero token costs
Perplexity has launched "Portable Computer" — a new agentic AI that runs locally on NVIDIA hardware and saves token costs.
Cloudflare teams up with big browsers to help websites tell welcome from unwelcome visitors
Makers of Chrome, Edge, Firefox back bot-fraud defense called Private Access Control Tokens
A Price War Is Brewing Between OpenAI and Anthropic — And It Could Make AI a Lot Cheaper for Your Business
OpenAI is weighing drastic cuts to what it charges for AI tokens. It expects Anthropic to do the same.
NVIDIA Launches DGX Spark 64 GB Desktop AI Systems
Local AI is becoming more useful by the token. As AI agents move from experiments into everyday development, increasingly capable open models are shrinking to fit on more devices, giving builders more to run locally. Coming this month, NVIDIA DGX Spark will be available with 64 GB of unified...
Will AMC’s New Leawood Films Distribution Arm and Tokenized Share Dispute Change AMC Entertainment’s (AMC) Narrative?
In August 2026, AMC Entertainment Holdings launched Leawood Films, a low-risk film distribution arm based in Los Angeles and Leawood, Kansas, designed to fill excess theatre capacity by focusing on fully financed small and mid-sized titles while respecting traditional PVOD and SVOD windows. Around the same time, CEO Adam Aron’s public clash with Robinhood over tokenized AMC shares highlighted rising tensions between traditional equity markets and emerging digital trading products, raising...
Satya Nadella has a message for Microsoft's tokenmaxxers
Microsoft's CEO said workers should use the right AI model for the job as companies look for ways to control rising AI costs.
Microsoft warns of sophisticated phishing campaign heavily targeting health care organizations
Microsoft Threat Intelligence is warning of a large scale, multistage phishing campaign that disproportionately targeted the health care sector, sending “code of conduct” themed emails to lure users into credential theft and token compromise.
NVIDIA’s V100, An 8-Year Old GPU, Now Sells for $100 and Crushes Modern Consumer Cards in AI LLM Workloads
The 8-year-old NVIDIA V100 outperformed the RTX 3060 and RX 7800 XT in AI LLMs, achieving 130 Tokens/s at only $200 total cost with mods.
Disney wants tech staffers to move fast with AI without 'tokenmaxxing'
Disney is leans into AI under CEO Josh D'Amaro but wants tech staffers to avoid "tokenmaxxing," or wasting tokens.
Meta's AI agent for WhatsApp Business is now available globally
WhatsApp will charge businesses for using its AI agent based on token usage
Datavault AI (NASDAQ: DVLT) Partners With Demora Foundation To Power K-Entertainment Tokenization Platform
Austin, Texas, United States, April 2nd, 2026, FinanceWireDatavault AI NASDAQ:DVLT, an AI-driven data monetization, RWA tokenization and Web 3.0 infrastructure company, and Demora Foundation announced a Technology Integration Agreement naming Datavault AI as the formal AI data intelligence and RWA…
AMC CEO Adam Aron blasts Robinhood for 'quasi-fake' tokenized stocks
"I hereby call on you and Robinhood to voluntarily CEASE AND DECIST the trading of AMC stock tokens," Adam Aron said.
'Tokenmaxxing' is the new Silicon Valley AI debate
Some engineers are "tokenmaxxing," or spending as many AI tokens as possible. But are leaderboards tracking token use the answer?
A $1,500 foundation model that rivals larger LLMs
Sapient researchers trained a 1B reasoning model on just 40B tokens — scoring competitively with 2B-7B models at a fraction of typical pretraining cost.
A note to readers: Science, accountability and elections
Dear readers, As early voting begins, I wanted to share some news — and why it connects to the decisions we make at the ballot box. I’ve been selected for the inaugural cohort of the South Atlantic Climate Journalism Hub mentorship program, hosted by Covering Climate Now. I’m excited to join this group, learn from scientists and fellow journalists, and expand my science reporting. Science directly ties to governance. Local officials make decisions about where development goes, how infrastructure is maintained, what emergency preparations receive funding, and how communities rebuild after disasters. Understanding the evidence behind those decisions helps us evaluate whether our government is serving us well. Here in western North Carolina, those connections are all around us, in Lake Lure and beyond. Flood maps inform decisions about property and development. Engineering assessments help identify risks at dams and bridges. Rainfall forecasts guide emergency preparations. Geological information helps explain landslide hazards. Water testing gives communities information they need to assess conditions. Meteorologists’ knowledge and public perception of weather events shape public safety. Each involves scientific or technical knowledge. Each also raises questions about public responsibility: What did officials know? How did they use that information? What did they tell the public? What did taxpayers receive for their money? A study can identify a problem. Elected officials still have to decide what to do about it, how to pay for it, and how to explain their choices. Reporting should examine both the evidence and the response. That is where I want to strengthen my work. My reporting on Helene recovery, infrastructure, flooding, public safety, and public spending has repeatedly brought me to questions that require scientific expertise alongside public records. This mentorship may help me better understand the research, ask more precise questions, and explain the answers clearly. I also bring the perspective of someone living with these concerns. I live downstream of Lake Lure Dam, and landslides affected my property during Helene. Those experiences bring urgency to my questions. Careful reporting helps establish the answers. Readers deserve to know what the evidence shows, where uncertainty remains, and whether officials accurately describe the information guiding their decisions. You also deserve scrutiny of proposed solutions: what they cost, what they are expected to accomplish, and what risks remain. Access to that information matters. As a mostly deaf journalist who relies on written communication, I will continue pressing for written public warnings, records, and explanations that people can read and use in real time. As you consider the candidates seeking your vote, consider how they approach evidence, preparedness, and transparency. Do they support local journalism, even when it’s not parroting a press release? Do they explain weather events clearly in plain English and explain their decisions after them? Make supporting records available? Acknowledge uncertainty? Follow through when a report identifies a risk? Those are practical questions about governing. Thank you for supporting Cops & Congress and the work that makes these questions part of the public conversation. I’m grateful to David Dickson and the CC Now team for this opportunity, and I look forward to bringing what I learn back to you. Annie DanceEditor & Publisher, Cops & Congress — Learn more about this newsletter, ethics policy, how you can help shape this work, and support it. Follow on X and Facebook. Send constructive criticism, fan mail, and tips with public documents for future stories: CopsandCongress@gmail.com ICYMI: View the Cops & Congress archive — Annie Dance is the publisher of Cops & Congress, a newsletter that analyzes what happens when crime, courts, disaster, democracy, and small-town policies collide. Views expressed here are covered by the First Amendment. Dance has a Bachelor of Arts from Manhattan University in Communication with a focus in Journalism and Government. She has been a journalist for over 20 years. Share Upgrade now or give a gift subscription
Sundar Pichai: Google’s Frontier Imperative, Compute Scarcity Dilemmas, Future AI Products and Businesses
Alphabet CEO Sundar Pichai has little patience for doubts about Google’s willingness or capacity to compete at the frontier of AI. His company’s latest Gemini 4 Argon model — long-promised and recently delivered — beats OpenAI and Anthropic’s latest public models on coding, math, science, and knowledge-work benchmarks and has placed Google’s AI back among the world’s best. Amid chatter that Google was fine with good-enough lightweight ‘Flash’ models, Pichai pushed back hard against any suggestion that company is limiting its research ambitions. “We won’t do it any other way,” he told me in a Big Technology Podcast interview recorded this week at Google’s Mountain View headquarters. Pursuing cutting-edge research is costly for Google, requiring compute and chips that otherwise could be sold at a high markup by its cloud business. But Pichai said allocating resources to frontier AI development is “ground zero” for Google. “Our cloud business is at any given moment in time forgoing short-term revenues because we are making compute trade-offs to support frontier AI development,” Pichai told me. In our conversation, we discussed Argon (which Pichai says processed three times as many tokens as Gemini 3 in the 72 hours after launch), AI’s role in developing better models, the AI interface of the future, competition with OpenAI, Google’s partnerships with Anthropic and Apple, data centers in space, and the nature of his optimism. You can read the full conversation, lightly edited for length and clarity, below, or listen to it on Apple Podcasts, Spotify, YouTube, or your podcast app of choice. Alex Kantrowitz: Google is back at the frontier with Gemini 4 Argon, which beats OpenAI’s and Anthropic’s public models on benchmarks spanning coding, math, science, and knowledge work. The conventional wisdom was that Google had given up on the frontier to focus on fast, good-enough models for its products. That was wrong? Sundar Pichai: Very wrong. Look, we are so committed to being at the frontier. It’s in our DNA. We won’t do it any other way. We’ve been thinking about the AI frontier for well over a decade now. The frontier is very dynamic. It’s very competitive. It’s very exciting. I couldn’t be more pleased with how Argon has come together. We took our time with it. We made sure we tested it safely and deployed it carefully. We took a bit longer than our other models, given its capabilities. Across the board, be it for agentic work, knowledge work, coding, creative tasks, Argon is one of the really good frontier models. The first 72 hours since we launched, it’s almost up to 25 trillion tokens used, which is about three times what Gemini 3.0 had in the first 72 hours. And we are capacity constrained as we are serving it. Excited, but there are always areas….. You know, the frontier is very dynamic, and you have to keep evolving the models. There are areas which we can see we need to improve as well. We’ll be at work on all of that as well. Which areas? On any given model, Argon would be across the board better at everything if you just look even two months back. But the space is so evolving. Even on coding, for example, there are pockets where we know we need to do better, and there’s clear feedback. We’ll make it better. The argument for standard models is that they’re already good enough to build great products, with the frontier valuable only for certain tasks. Meta built Muse with good models rather than the best, and Google could put standard models into Maps, Gmail, and Search. Why isn’t that sufficient? It’s definitely not how I have thought about it at all. We are deeply committed to driving the state of the art and actually pushing where this technology can go, all the way from transformers. That’s a journey we’ve been on. You will see us continue to do that. When you’re at the frontier, you get an earlier look into the emerging capabilities, earlier than others, which helps you better prepare for the future. We are fortunate to have a wide set of opportunities across consumer and enterprise. Some of our cloud customers, for sure, want the model at the frontier. The frontier models will be better at developing the next set of models in an era of recursive self-improvement. I would argue it matters there. Internally, we have definitely seen a steep point of inflection by deploying Argon to our developers. It definitely tackles problems which our previous models could not do. Being at the frontier puts you at the innovation frontier and helps you drive that as well. There is room for models of all sizes and capabilities. That’s why we did Flash, Flash-Lite, etc. I agree with you that you can develop very good product experiences with not necessarily the most frontier model, because they tend to be slower and more expensive too. But we care about the Pareto frontier, and we’re committed to doing all of that. You said developing frontier models gives you a look at what’s coming. What do you see on the horizon? What makes this moment so exciting is you build something which you think is a culmination of months of work, and you put it out. You already see clear headroom ahead. We’ve all been working on these agentic, long-horizon capabilities, but clearly the frontier is going to have multiple of these agents working together to do things. That’s an example of an exciting paradigm underway. It’s also clear to me that users, both in consumer and enterprise, will have access to some multi-agent framework for them, working all the time doing things. It feels like a moment of infinite possibilities. Sergey Brin said one of Argon’s core innovations was relying more on the model to help develop the model. Yet incidents involving OpenAI’s bots breaking containment and Google’s bots accessing a site they shouldn’t have suggest these systems still aren’t perfectly aligned. Isn’t there reasonable cause for concern about accelerating toward recursive self-improvement? There are three parts to your question. Let’s first talk about RSI itself. You’re right, Sergey mentioned RSI being a critical contributor as we are developing Argon. But to be very clear, today, in a continuum, we are pretty far from the end state of RSI or something like that. The right way to describe it is you have RSI loops along with AI researchers in various aspects of model development. You almost have agents helping AI researchers in some automated way work together to build these models. It’s good to at least understand where we are today, and it’s super exciting. I expect that continuum to keep making progress as we go into next year. The second question is around safety. Security comes from defense in depth. You’re talking about alignment, obviously. We all are doing a lot of work to make sure the models are as aligned as possible. But it’s an active area of research, and there are clear gaps. That’s not the only thing we rely on. You have the work you do on alignment, but beyond that, we put a lot of effort into real-time monitoring as these models are working. Making sure you have good telemetry into what’s happening so that you can intervene in real time if you see unexpected behaviors. How you do that, the architecture with which you do that, how you make sure it’s all sandboxed and isolated in a way that it can continue doing its job during real-time monitoring — these are all areas of investment. Obviously, when we deploy these models, giving customers, enterprises, control, both with real-time monitoring, but their ability to decide where they deploy, what type of data which models ever have access to, etc. All that work is very, very important, and we are doing a lot of that. As an industry, we need to constantly up our game and take the time to do all this stuff, including dedicating the compute required to make sure you can safely deploy these models. The third part is: Is it all enough? This is why there are a lot of conversations in the industry, and I think it’s reasonable to be worried. The right way to approach something you worry about is early on thinking about what more we can do. As an industry, you’re definitely seeing increased levels of work on alignment, testing for safety, etc., as we deploy these models. The next step would be developing better standards so there are cross-industry standards of how we do these things, including the ability to evaluate using independent evaluators. If you look at areas like financial controls, we have accounting standards, we have independent auditors. Those are all the next natural steps for us as an industry to work toward, and regulation will end up playing a role as we go through all of that as well. Google pursues both language models and world models, while some other AI companies like OpenAI have focused more heavily on LLMs as the path to AGI. Why is world-model research important to you? As humans, you obviously perceive the world around you, and you’re taking all those inputs. Making sure these models are able to understand the laws of physics, simulate the world in high fidelity. All that, I think, are key enabling blocks for intelligence. We are working across all possible applications of this. Take an area like robotics. It’s not an accident: If you look at embodied reasoning benchmarks, Argon is state of the art. That all comes from our experience building world models. Inherently, I would expect there will be architectures which will converge all of this in the future. But for us, we view this as, today, this jagged frontier that is intelligence. We want to be at the frontier of all the possible approaches, and that will help us both drive better research, better insights, and develop a much better AGI model, is how we think about it. Some people argue Google should focus on selling its computing resources rather than using them for AI research. Cloud wants to sell compute at a high markup, while researchers need a very big computer to build the best models. How do you balance those demands? Ground zero for us is making sure we are investing in the compute that’s needed to be at the AI frontier. In 2015, I set the company to be AI-first. AI is that key horizontal ingredient across everything we do, all the way from Search to Cloud to Waymo. That is ground zero for us, making sure both for research and developing the models that we have enough compute needed. Beyond that, we have these independent businesses which each have their own long-range plans. They’re using those long-range plans to project forward, and we plan for compute for all of them. The cloud team, they have a business and they have a projection and they have demand signals. They have committed contracts. With that help, we can plan on a long-term compute trajectory for them. We’re fortunate to have this wide set of opportunities. There’s no company in the world which isn’t making trade-offs at any given moment in time. But the first principles we operate on is making sure we have enough compute to invest in the frontier, and that’s non-negotiable, and everything else works on top. Remember, we are building our own compute, our own data centers. There are long lead times. The compute we have today is a function of decisions I made in 2024. Even if you think back to a year from today, last year the agentic coding takeoff hadn’t happened. Today, obviously, when you sit and look at anything, everyone is short—not just us, other companies—and everyone wishes you had two times what you had planned for. That comes with the territory. At any given moment in time, we are planning for what we would need two to three years out in the future. You’re going to find some decisions you make may look suboptimal two years out, depending on where the state of the art is. I don’t spend too much time thinking about it. The way I think about compute allocation is making sure Google DeepMind, both for research and frontier model development, is cutting edge in terms of compute access. After that, we plan for all our businesses. It’s great we can provide our compute to other people too, like we have done. It allows us to scale that up, get real tests for it in the widest application set possible, test it in the real world, and make sure we are at the cutting edge of technology at every layer of our AI stack as well. Ben Horowitz relayed on a podcast that Andy Jassy had said compute was sold out until 2028. Are you seeing the same thing? If today a customer is coming out and asking us for compute in ’27, that is incredibly difficult to do. Even last year we went out and leased some short-term compute to accommodate some customers and help them out. But yes, we are all dealing with similar timelines. Every major technology shift seems to involve overbuilding, from railroads to fiber. Do you think that’s happening here? It depends on the time frames you’re looking at. This level of build, if I take a long enough time frame, it makes complete sense to me. We are at early stages of the most consequential technology ever, and this is not like fiber. It’s not dark. We are using it. We are capacity constrained on a daily basis. The applications, the underlying technology is improving rapidly. But running a company, the way I work through the moment is you look at your demand signals. You look at your committed revenue opportunities that you’re getting in the form of contracts. You balance that with the underlying technical capability, or internally you balance how we are using it internally, the model trajectories. Based on all that, you take a disciplined approach—a confident, aggressive, but disciplined approach—to plan out compute. But when you talk about the broader economy as a whole, you’re having people operate with various levels of risk. I focus on what I can control as a company and plan through it. If your question is, can there be an overbuild? Yes, because you can have some constrained components. Remember, what matters is a set of everything you need to build out that compute. Memory, electricity, electricians…. Land. Like 2,000 components or more. You could get constrained at some point. There could be things which cause bottlenecks, and then you could have overbuilds. There could be external shocks to the economy. There are macro risks. All that is possible at any given time. But am I confident that the compute we are planning for, we will have uses for it given enough time? Absolutely, yes. I was in the room when you said AI would be as profound as fire and electricity. If you believe that, shouldn’t you give researchers the biggest possible computer, even at the expense of today’s cloud profits? We’re all doing some form of that. Our cloud business is at any given moment in time forgoing short-term revenues because we are making compute trade-offs to support frontier AI development. What you’re seeing is true already. But you want to be thoughtful. Even our AI researchers have some constraints, which are also good. That’s what inspires creativity. That’s what helps you come up with a more efficient algorithmic breakthrough, maybe to use less memory. What we are committed to is to invest what’s needed to stay at the frontier. OpenAI President Greg Brockman told me a few months ago that the ideal AI product might have no interface at all—just as natural as conversation. What do you think the interface looks like in the age of AI? Computing historically has been humans adapting to computing. With AI, it’ll be the other way around. The technology will adapt seamlessly to how humans live and go about their day-to-day life. You want to be able to speak to that AI when you want to. It has context. It can be persistently running somewhere, give you continuity. You’re not dealing with a lot of complexity. It’s natural. All that makes sense to me. That doesn’t mean it won’t have any interface, however. Nowadays, even on my phone, Gemini voice models are so good. I can just press the microphone button. I can just ramble. That’s why we call it Rambler. I can ramble. It intelligently corrects with my personal context. Even my friends’ names, weird names, it gets it right, even if it first guesses it wrong. It’s amazing. I’m not going in, backspacing, editing once I finish saying it. It’s remarkable how good it’s gotten. But there are going to be moments you will need an interface. You may want to entertain yourself. You may want to watch a video. You may ask your AI to create something, and you want to watch it, consume it. The way I think about it is, you want it to be as seamless and as natural as possible, and that you are not doing the hard work adapting to whatever the interface is. The technology is adapting to how you’re working at any given moment in time. When I asked Sam Altman about competition from Google, he argued that OpenAI was building AI-native products and implied Google was bolting AI onto existing ones. Do you accept that framing, and can a “bolt-on” compete with an AI-native product? First of all, it’s a very dynamic moment. We are all pushing each other. OpenAI has pushed us as a company too. This is not just we are doing something to them. Far from it. Yes, Google went into its own version of a code red too. In the AI moment, all of us have to respond with varying levels of urgency at any given moment in time. We are constantly building new products as well. Gemini was built from the ground up as a product—the Gemini app I’m talking about. Or tomorrow, as all of us are adapting to agents, are you bolting agents onto ChatGPT, or are they building it natively? How would you think about that? You can always put these questions back. I don’t think of it that way. I think of it as you want to build world-class product experiences that are AI-native and they are winning. Some of it is adapting products, some of it is rethinking new experiences. They will all be part of how you approach this moment. How do you expect to make money from these AI interfaces? If an agent recommends a product and I buy it, should it take a cut of the sale, even if that’s disclosed? I’m going to put enterprise aside. Consumers are going to increasingly value this technology. Take software developers today. Developers are willing to pay for these coding models because it really helps make them more productive. Nobody would want to code the way they were coding three years ago, or even the same time last year. When you’re delivering value, there’s value to be had. I’ve never seen examples where consumer products really scale to billions of users, where people are using it intensely, and haven’t had monetization to go along with it. If you’re providing value to people, there is inherent value to be had. The nature of it may change. You’re asking a scenario in which agents are doing stuff for you. At the end of the day, you will have commercial needs. The agents need to fulfill those commercial needs. The providers on the other side are going to see value in that. There’s going to be some mechanism by which it gets fleshed out. That’s what an efficient system does. But what matters is that you’re building those world-class consumer experiences. I promised myself one product feature request: Can you make Gemini-powered search in Gmail as good as searching Gmail through ChatGPT? What’s going on here? One of the exciting things we announced recently is the unified Gemini agent for enterprise, and we’re bringing that to consumers. These are persistent agents which run 24/7 in the cloud. We’re spinning up a micro VM effectively for every query, and from the runtime, the harness, the model—it’s all a headless service. It’ll work everywhere, and it has your context with you. You can address this agent whether you’re in Workspace, in Search, in Gemini. Agents are going to drive this unification from a customer experience, which people are looking for. When you have a broad product portfolio and you’re moving forward fast, there are always areas where you’re moving forward faster than some other areas. That’s valid feedback. But we are putting a lot of time and effort into making sure, as we build out our agentic experience, that unification and that capability works well across all these services. People have asked where Google’s equivalent of Muse or Dots is. Do you anticipate creating a new product to compete with them? Our goal with Gemini is to be a personal, proactive, and powerful assistant. That’s what we are working toward. You saw us describe the early versions of it at Google I/O this year with Spark. Effectively, already, if you open the Gemini app as a Pro or an Ultra user, you see Chat and Task, and they start giving you that agentic experience. With Argon now, that capability is dramatically accelerated. What we announced recently in our enterprise event, we are bringing that to consumers as well. Soon in Gemini, you will be able to ask something, and it’s a continuum. You don’t need to think about whether you’re asking a chat or an agentic task. It’ll just automatically do the right thing, and it can work persistently in the background securely and get things done for you. Already, empirically, when I look at what’s coming into Gemini in our app, last year you would have tagged them and most of them would have been chats. But a steadily growing percentage is already agentic tasks in Gemini. Yes, we are going to bring personal agents to our consumers. It’ll work in the Gemini app. It’ll work in Google Search. It’ll work in Gmail. You can address it, call it, etc. We want to make sure we’re getting it right. It’ll be powered by Argon, and we are very excited to bring it to our users. Google has invested and committed tens of billions of dollars to Anthropic. What do you see in Dario Amodei that makes you want such a deep investment and partnership? Very early on, when we looked at the founding team of Anthropic, it was very clear to us this was a world-class team which was involved in some of the earliest GPT models. We were quickly able to assess that they would be at the frontier. We’ve always tried to identify and partner with companies which would be at the cutting edge of technology. We’ve done that with SpaceX. We’ve done that with Anthropic. That’s what we saw, and we had a chance to partner. Is your partnership with Apple to license Gemini for Siri expanding, contracting, or standing still? For both us and Apple, that partnership was about translating the capabilities of AI into better experiences for users. I thought Apple has done a phenomenal job of incorporating—they customize these models and build it into Siri—and it’s been great to see the reception. We are committed to doing more. You’re going to do more with them? There’s nothing specific to comment, but it’s a partnership. It’s working well. What have you learned so far about the feasibility of data centers in space? It’s a very exciting project. Projects like this remind me—I still recall that moment in the desert for the DARPA Grand Challenge, watching the Stanley car compete in it, which ended up being the basis for Waymo. With all these technologies, you’re at a moment in time where it looks a bit like sci-fi, but 10 to 20 years down the line, it’s actually working in a meaningful way. I feel the same in space. We’ve always believed in betting big on these technologies. Quantum has been a great example of it. We’re probably 10 years into the journey. Waymo, we are almost 20 years into the journey. Space, we are two to three years into the journey. It’s clear to me that it’s an approach which will make a lot of sense in the future. Being in sub-synchronous orbits with 24/7 access to the most abundant form of energy we have access to will make a lot of sense for this exponential increase in compute that we are going to need. Having said that, this is a challenging project. The team has disciplined milestones. It was a super exciting milestone to send the TPUs up to space. Sometime, as we are speaking, we are expecting a “hello world” moment from them. I look forward to seeing that happen, and we’ll continue investing here too. You previously said 25% of the code written within Google was written by AI. What’s the number today? That number in our world today is meaningless in the sense that every part of what we now touch is influenced by AI deeply. What we are tracking with Argon, for example, it’s been a point of inflection. People are using Argon to really go tackle complex projects, including projects which affect millions of lines of code. In the four weeks since we have launched Argon, we have seen a 30% increase in CLs, and more importantly, almost a similar increase—a one-third decrease in rollbacks. People are writing code more productively and better code with it. We are tracking it with newer metrics now. How are you dealing with opposition to data centers? If you can’t build in the U.S. the way you want to, where else do you go? We’re very committed to building our data centers very responsibly. When I look at our largest data centers in some of the communities where we have been the longest, I visit them, I talk to people. There are people who come and tell us their kids have gone to college because of our investments in these communities, and we’ve grown in those communities. To me, this is about doing it right. Engaging with the communities early, doing it transparently. We are committed to making sure we are bringing incremental power, and we cover 100% of the cost of our power needs for these projects. It’s understanding the impact on neighborhoods and making sure you’re working thoroughly to address them. Done rightly, you can continue making progress. But we are constantly taking feedback into account. I do worry not everyone operates the same way. People have been surprised by data centers showing up in their backyard. Those are all important. As an industry, we should continue to do better. But part of the reason, in a 20-year time frame, we are thinking about things like data centers in space is making sure we are able to put these data centers where maybe the trade-offs will be different. What unproven area of AI is Google working hardest to crack? Is it continual learning—getting models to keep updating after release—or something else? There are multiple parallel paths. You saw with Argon we have a million output tokens. With our Gemini models early on, we had a one-million context window on the input side. These are all examples of breakthroughs we are constantly working on. How do you think about synthetic data? How do you think about continual learning? How do you think about more efficient architectures? We do dedicate a meaningful percentage of our compute to these research bets too. It always goes hand in hand, including our world models. Those are all bets we are placing. Sometimes, as part of these bets, there are ideas that emerge which just make their way into the mainline models. Maybe we don’t fully talk about it, but we have multiple steps in these areas. Working in technology requires optimism that building something will improve the world, but those bets can leave real consequences when they fail. I recently returned to a town in El Salvador where enthusiasm for Bitcoin had given way to unfinished construction and something resembling a ghost town. How do you maintain the optimism you spoke about in your Stanford commencement address while accounting for those risks? I genuinely have a deep conviction that improving technology and improving access to technology makes people’s lives better. It’s been my personal experience in my life, out of which comes that deep conviction, optimism. But with that, I’ve always felt humanity’s journey is always about working hard to harness that technology to make it work well for society. The optimism is not an optimism that you just build whatever you want, whenever. You work hard to make sure you’re building technology in the service of humanity, to make the human condition better. That’s got to be the defining goal as we are approaching it. With that framework, I’m very, very optimistic. We are in very early stages of the kind of progress that we are going to see on many areas. We’re all talking about areas like healthcare. A lot of the progress is ahead. The ability to make things in a fundamentally cheaper way on many things where things have been getting expensive for a long, long time, discovering new materials, etc., which really bend the curve—it’s tough to step today and understand the impact of all that, but they’re going to be very profound. I’m very optimistic, but we have to be rightfully concerned about all the ways this technology can go wrong. Sundar, I know you don’t do many of these. Thank you. Thanks, Alex. Appreciate it.
Russia’s SVR borks public Wi-Fis for digital surveillance
Keyloggers, audio-visual surveillance, and token theft on CaptivePortal's agenda as hospitality sector put on alert
FBI warns of Kali365 phishing service targeting Microsoft 365 accounts
The FBI is warning about the Kali365 phishing-as-a-service platform (PhaaS) that is used to hijack Microsoft 365 accounts by abusing OAuth device code authentication to steal session tokens and bypass multi-factor authentication (MFA).
Amazon shuts down AI leaderboard after employee 'tokenmaxxing'
Amazon nixed an employee-created AI leaderboard called "KiroRank" after concerns it encouraged excessive AI spending.
Nvidia launches free tool that links idle computers into a personal AI data center
We have AI tokens at home.
Internet Restrictions Spur Russians to Openly Question Putin’s Moves
From beauty influencers to the token political opposition, Russians are openly questioning President Vladimir V. Putin’s moves to hamstring access.
Alibaba CEO Takes Direct Control of New AI Business Unit
Alibaba Group has formed a new business unit called Token Foundry as part of the company’s reorganization plan to strengthen its artificial intelligence
Marc Benioff said Salesforce will spend $300M on Anthropic tokens
Salesforce CEO Marc Benioff said he's able to work faster with AI agents and that the company is working on tech to make coding easier at Slack.
CineNow Taps Digital Asset Executives as Entertainment IP Fund Targets $150M and Tokenization
Nikita Sachdev joins as Advisor and Hamed Faqirzai as Director of Tokenization & Capital Markets to help shape CineNow’s tokenization strategy. ...
Tokenized Shares Dispute Could Reshape The Case For AMC Stock
AMC Entertainment Holdings recently objected to Robinhood offering tokenized versions of AMC shares without company approval, while Robinhood and Coinbase both moved to attach traditional shareholder rights, including voting and redemptions, to their tokenized stock products. The clash over tokenized AMC shares raises practical questions for investors about who truly controls voting power, how ownership is recorded, and what that might mean for a heavily scrutinized balance sheet and ongoing...
Datavault AI Inc. (NASDAQ: DVLT) and Demora Foundation Execute Technology Integration Agreement to Power the K-Entertainment & K-Wave Global Platform
DVLT × Demora Foundation: Datavault AI's Enterprise AI Stack Deployed as the Data Intelligence & RWA Tokenization Layer for K-Pop,…...
What Are Companies Getting for All That A.I. Spending?
A new field of “tokenomics” has emerged to measure the return on all the money companies are pouring into artificial intelligence.
(BPRW) SRT Entertainment Revives Knock’em Out The Box Records With 001myke’s Debut Album “SEVENTEEN” - “Picture Perfect” Official Music Video - Submitted for GRAMMY® Consideration
(Black PR Wire) SRT Entertainment has revived Knock’em Out The Box Records, the New York hip-hop label founded in 1989 by veteran record executive D.A. Johnson whose career includes Bellmark/Life, Malaco, eOne and C&C Music Factor...
(BPRW) USHER'S NEW LOOK, BET AND BOYS & GIRLS CLUBS OF GREATER DETROIT PARTNER TO EXPAND PATHWAYS AND DEVELOP THE NEXT GENERATION OF ENTERTAINMENT LEADERS THROUGH BET NEXT GEN PROGRAM
(Black PR Wire) New York, NY - Usher's New Look, Boys & Girls Clubs of Greater Detroit and BET Next Gen, BET's leadership development initiative, today announced a collaboration designed to expand access to careers in media, music...
Today's NYT Strands Hints (and Answer) for Sunday, October 4, 2026
If you’re looking for hints and answers for Sunday, October 4, 2026 with the theme “Grab your tokens,” read on.
Google won the search war. It's using the same tactic to win in AI.
Inference costs are becoming a race to the bottom, and Google's full stack AI approach puts it in a good position to offer cheaper tokens.
Nvidia's Jensen Huang pitches tokens as recruitment strategy
Jensen Huang said engineers will soon be asking "how many tokens comes along with my job?" and is willing to pay them tokens worth half their salary.
Clawdmeter – A DIY ESP32-S3 desk dashboard for Claude Code token usage monitoring
Clawdmeter is a DIY ESP32-S3-powered desk dashboard that displays Claude Code token usage on a 2.16-inch AMOLED screen so you know when you're about to
Securitize Completes Business Combination with Cantor Equity Partners II
/PRNewswire/ -- Securitize Corp. ("Securitize"), the world's leading tokenization platform with over $4 billion in assets brought onchain, today announced the...
Solidity Pro VS Code Extensions Steal Crypto Wallets, API Keys, and Credentials
Malicious Solidity Pro VS Code extensions steal crypto wallets, API keys, SSH keys, and developer tokens, then exfiltrate the data via Telegram.
AI token Hunger Games are coming for software engineers at work
AI token spending soared from companies pushing unlimited use. Now tech workers are competing for compute as companies rethink budgets.
Gemini 4 Argon Arrives With Big Claims for AI Agents and Scientific Discovery
Google just released the new Gemini 4 Argon, which it claims can generate up to a million tokens in a single run, a big jump for long code migrations and multi-step agent work. Competitors top out around 128,000 output tokens per request, so based on that claim, Argon’s 1 million is roughly 8x higher. Argon […]
Why politicians should pay more attention to the polls (reprise)
Welcome to the 231st edition of The Week in Polls (TWIP), a polling newsletter that always welcomes new readers. So please don’t keep TWIP a secret; share it with your friends and colleagues: Share The Week in Polls With all the stories swirling around Reform, JL Partners and political polling, I am finally tearing myself away from talking about the regulation of polling this week - and instead re-running a previous piece reminding us why the solution to problems with polls isn’t simply to ban or ignore them. Rather, we need them to play a vital role in making our politics - and politicians - better. This is followed by summaries of the latest national voting intention polls, seat projections from MRPs and the most recent party leader ratings. Subscribers who pay for the full edition can also read ten insights from the latest polls and analyses, including: A trio of organisations and a pair of professors have been casting their eyes over the Liberal Democrats ahead of the party’s autumn conference; What impact did the summer heatwave have on views on climate change? and Which wins out between chocolate and cheese? If you are not yet a paid subscriber, you can read all ten by starting a free trial now: Get FREE 7-day trial For more polling news as it comes in ahead of next week’s edition, you can also follow The Week in Polls on Bluesky. Picking up on my doubt last week about whether or not 35% of British train travellers have tried to order a takeaway for delivery to a railway station or carriage, thank you to reader Iain, who has been in touch to say he has done just that… in India. Another quick pair of updates, this time on the controversies around Reform and polling. Politico has a further scoop - “Farage aide caught in donor sting attended Egypt retreat with JL Partners” - and Ian Dunt has written about the problems with polling too: “Seeing the response of many polling experts in the last few weeks suggests that they are nervous about what is happening and keen to mitigate against it. Seeing the response of journalists, which has been basically non-existent, suggests the precise opposite.” My usual reminder as well: please do hit reply if you’ve spotted a poll, focus group or new piece of poll-based research that you think should be in contention for coverage in a future edition. Thank you! And with that, on with the show. How politicians get public opinion wrong I have written before about the benefits that come from widespread availability of high-quality political polling. We need more polls and more good coverage of polls. However, what I’ve not dug into before is the research into how accurately politicians estimate public opinion on issues. Thankfully we have academic research to tell us. First, the bad news: We present results from a study of 866 politicians in four countries.1 Politicians were asked to estimate the percentage of public support for various policy proposals. Comparing more than 10,000 estimations with actual levels of public support, we conclude that politicians are quite inaccurate estimators of people’s preferences. They make large errors and even regularly misperceive what a majority of the voters wants. Politicians are hardly better at estimating public preferences than ordinary citizens. Walgrave, S., Jansen, A., Sevenans, J., Soontjens, K., Pilet, J.B., Brack, N., Varone, F., Helfer, L., Vliegenthart, R., van der Meer, T., Breunig, C., Bailer, S., Sheffer, L. & Loewen, P.J. (2023a), “Inaccurate Politicians: Elected Representatives’ Estimations of Public Opinion in Four Countries”, The Journal of Politics, 85(1): 209-222. The bad news is not quite as bad as that makes it sound because, when asked whether the majority of public opinion was for or against various policies, politicians picked the wrong option ‘only’ 29% of the time. However, this included getting it wrong even when public opinion is far from evenly split on an issue: They do not only get it wrong when the public is divided and hard to read (e.g., when there is a 51%:49% distribution). Seventy percent of the misplacements occur for policies with a relatively clear distribution of at least 60% (dis)agreeing citizens. Moreover, the public scored only slightly worse than politicians, picking the wrong option 2.7 out of 8 times on average, compared with 2.3 times for politicians. Turning to the other measure of accuracy (not which side is in the lead but what the percentage support is for each side), politicians on average were 18 percentage points out in their guesses. Again, this was only slightly better than the public, whose error came in at 21 percentage points.2 Further research, using the same mix of countries, gives a little more comfort about how (in)accurate perceptions are: We show that politicians have a better understanding of public opinion when they think the issue matters to voters. Further, when an issue is personally important to politicians they more accurately estimate their party supporters’ opinions. The results confirm that politicians hold more accurate perceptions of voters’ preferences when they think it is important to do so but not necessarily when the issues actually are important to voters. Butler, C., Walgrave, S., Soontjens, K. & Loewen, P.J. (2024), “Politicians are better at estimating public opinion when they think it is more salient”, Party Politics: 1–13. Even so, it’s not a good picture. The remedy for this? Politicians should spend more time reading the polls. So forward this newsletter to someone you know… Share For a more general argument against banning polls, see my 2024 piece. Is “Let Bartlet Be Bartlet” good advice or not? The latest episode of my Political Fictions podcast heads across the Atlantic for a dip into one of the political TV classics: Mark and Cory talk about a giant of American political drama for the first time this week, as they tackle The West Wing. They talk about Let Bartlet Be Bartlet, the nineteenth episode of series one. In it, Mark wonders if The West Wing is actually the epitome of the “Boris Johnson school of politics”, Cory outmanoeuvres Mark into confirming we will record a Christmas Special about Love, Actually and they debate whether the central message of this episode has been misunderstood. Cory’s email newsletter is Paperback Rioter and Mark has a family of email newsletters. Subscribe to our newsletter for some additional bonus content and to get alerted whenever a new episode comes out. Our theme tune is “Monkeys Spinning Monkeys” by Kevin MacLeod/incompetech.com and licensed under the Creative Commons: By Attribution 4.0 License. You can also listen to our episode on Apple Podcasts, Spotify, the web or YouTube. Elsewhere from me… [ A Lord's Eye View How to fix the problem of money in our politics Welcome to my latest update on work in the House of Lords. This time I am taking a look at how our political system is regulated, and the protections we need to keep democracy safe from becoming simply a contest to see who has the biggest bank account… Read more 5 days ago · 3 likes · 2 comments · Mark Pack ](https://lordseyeview.substack.com/p/how-to-fix-the-problem-of-money-in?utm_source=substack&utm_campaign=post_embed&utm_medium=web&embedding_publication_id=1269283) If you spot a factual error… Borrowing an idea from Stuart Ritchie and others, if you spot a factual error in an edition of this newsletter, I will give you a free 12-month subscription to the paid-for version (or an extra 12 months on your existing one). Factual errors qualify, but grammatical errors or disagreements over interpretation do not, and my decision is final. Messages about both of those are always welcome, though. Voting intention and leadership ratings Here are the latest voting intention polls from different pollsters. It is all pretty static at the moment, with a small and consistent Labour lead in nearly all the polls. The table is also online here, along with explanations about what is included. It is updated regularly throughout the week as new polls are released. Next are the seat projections from MRPs, also sorted by fieldwork dates, listing those which are based on dedicated fieldwork (rather than models based on averages of other polls). Because such MRPs are published infrequently, some of the entries below are quite old. Finally, here is a summary of leadership ratings, with a graph from PollCheck: Detailed numbers from the latest polls are online here, and both that and the graph are updated regularly throughout the week as new data comes out. For historical figures on both voting intention and leader approval stretching back to the 1930s, including Parliamentary by-election polls, see PollBase. Catch up on the previous two editions [ ](https://theweekinpolls.substack.com/p/a-crisis-point-for-political-polling) [ A crisis point for political polling Sep 13 ](https://theweekinpolls.substack.com/p/a-crisis-point-for-political-polling) [ Read full story ](https://theweekinpolls.substack.com/p/a-crisis-point-for-political-polling) [ ](https://theweekinpolls.substack.com/p/extra-questions-about-the-polls-in) [ Extra questions about the polls in Reform’s funding scandal Sep 6 ](https://theweekinpolls.substack.com/p/extra-questions-about-the-polls-in) [ Read full story ](https://theweekinpolls.substack.com/p/extra-questions-about-the-polls-in) My privacy policy and other legal information are available here. Links to purchase books or other items are usually affiliate links which pay a commission for each sale. Quotes from social media messages are occasionally lightly edited for punctuation and clarity. If you are subscribed to other email lists of mine, please note that unsubscribing from this one will not automatically remove you from the others. If you wish to be removed from all my lists, simply reply to this email to let me know. A quintet of Lib Dem polling analyses, and 9 other polling insights you shouldn’t miss The following ten findings from the most recent polls and analyses are only for subscribers who pay for the full edition, but you can sign up for a free trial to read them straight away. Ahead of the start of Liberal Democrat conference, we have a trio of organisations and a pair of professors with pieces out about Ed Davey’s party:
Engrams Embedding Entendre: Codesign for Efficient DRAM/SSD Offloading
Engram extends standard token embeddings with learned multi-token lookups. Recurring local patterns retrieve vectors directly, reducing the need to reconstruct them through attention and feed-forward layers. With Engram model architecture optimization, it allows for lower HBM capacity to be needed for models at the same quality. This does not mean there won’t be an insane demand for HBM but it just means that model architecture will continue to innovate around constraints. This model architecture design is naturally codesigned for parameter offloading: each token accesses a few embedding rows whose addresses depend on token IDs, not hidden states. The runtime can prefetch those rows from host DRAM while earlier layers compute, keeping the table outside HBM without transferring entire weight matrices. Our Memory model contains our latest estimates of quarter by quarter HBM, DRAM, & NAND supply and demand. Offloading frees HBM for model weights and KV cache, potentially supporting larger batches or more concurrent sessions. When DRAM becomes the next constraint, NVMe offers another tier. Recommendation systems already cache frequently or recently accessed embedding rows in faster memory while backing colder rows with SSDs. After NVIDIA roadmap had to change due to massively despec’ing Rubin Ultra from 1024GB to now ~200GB of HBM per chip, model architecture optimizations like emgram maybe helpful. Our DeepSeek-V4.1-Flash configuration uses roughly 189 GiB of memory for Engram. We replace it with a memory-mapped (mmap) file and measure serving performance with offload to SSD. Later on in our report, we will show our Engram offloading experiments along with the official InferenceX agentic inference serving results on Engram models like DeepSeekv4.1 Flash across all 6 NVIDIA GPU SKUs along with MI355X. Unsurprisingly, the CUDA Moat is still mogging MI355X on the ultra popular DeepSeekV4.1 Flash model. We also show how even on high capacity HBM SKUs, offloading emgrams to DRAM could result in even better performance for most of the pareto than keeping the emgram in HBM. Our benchmark has been widely reproduced, validated and/or supported by almost every major buyer of compute from Google Cloud to Microsoft Azure to Oracle, to Meta and many more. Furthermore, it has the support of the ML community including from vLLM, LMCache, SGLang, PyTorch, Huggingface and the support of major labs like OpenAI, MiniMax, ZAI, Qwen, Moonshot Kimi, etc. Star the InferenceX GitHub repository if you find the open-source benchmark and data useful!. InferenceX is the only inference benchmark in the world to have TPUv7, Jalapeño, Nvidia Rubin NVL72, AMD, and soon, SambaNova and Trainium. Due to how realistic AgentX scenario is to real world agentic inference workloads, AMD has committed to collaborating on MI455X UALoE72 too. DeepSeek did not release the original paper’s two trained Engram models. We replicated its setup on fineweb-edu using the released code and training hyperparameters, at an estimated 6E18 FLOPs per run. We observed the same U-shape scaling: Engram improved performance over pure MoE baselines. We also reproduced DeepSeek’s results, where earlier-layer representations with Engram resembled those of later layers. Like the original Engram paper, it is possible to probe Engram’s gate scores to see what n-grams DeepSeek-V4.1-Flash makes the most use of. Our gate scan finds names, code fragments, relational phrasing, and boilerplate. These examples prioritize interesting-ness over gate strength. One unexpected result was Wright : Ace Attorney. These examples suggest learned memory optimizes the training objective, not a judgment of which facts deserve storage. Licenses, bibliography fragments, API scaffolding, and website furniture can provide prediction shortcuts, so the value of additional Engram capacity may depend on what survives data preparation. This does not show that table capacity is “wasted”: the evaluation-corpus scan establishes neither training exposure nor the capacity occupied by each category. For offloading, strong gates do not identify cache-hot rows. Low gates do not automatically save reads either: computing the gate requires the retrieved key and undoes the performance gain of a fused kernel. Skipping reads would require a separate usefulness predictor before retrieval. In the original paper’s inference-time ablation, factual-knowledge benchmarks retained just 29–44% of their original performance, while reading comprehension retained 81–93%. This is due to the training–inference mismatch. The resulting degradation therefore measures this trained model’s dependence on Engram, not the performance difference between models trained with and without it. In our ablations, suppressing Engram worsens token likelihood across all evaluated domains, especially encyclopedia text and several code corpora. Surprisingly, GSM8K accuracy stays within measured run-to-run variation and removing Engram has no effect. Engram is not a detachable dictionary beside an unchanged MoE. Removing it changes downstream features and expert selection. We tested whether rerouting hurts or compensates by holding tokens fixed in a teacher-forced experiment on CRUXEval, a code-reasoning benchmark of small Python functions where the model predicts a function’s output from its code and an input, and scoring the reference answer. Removing Engram raised answer loss from 0.2848 to 0.3093 bits/token. Forcing the ablated model to use the original Engram-on expert choices made it worse still, at 0.3375 bits/token. Rerouting partially compensates for the missing memory. Memory features and expert selection work together, rather than following a clean “memory stores facts; experts reason” division. On the same CRUXeval, removing Engram during either phase reduced accuracy and increased generated tokens; removing it throughout produced the largest changes. Keeping Engram for prefill leads to more correct answers than keeping for decode likely due to semantically richer KV cache transferred to decode workers, allowing it to mitigate some of the performance loss. Engram’s table is large, but each lookup is small. DeepSeek-V4.1-Flash requests 24 rows at each of two Engram layers, about 12.4 KiB per processed token position across the model, or 3.1 KiB per GPU when split across four GPUs. Currently as of Day 7 since Model Release, MI355X is still 2-4x worse performance per dollar compared to B200 even when normalized by Mi355X’s lower TCO. Our full total cost of ownership breakdown comes from our AI Cloud TCO Model along with monthly market surveys of over 100+ gpu clouds & gpu cloud customers. On the Day 0 release of DeepSeekv4.1 Flash, NVIDIA vLLM works out of the box with zero issues across all 6 SKUs: H100, H200, B200, B300, GB200, GB300! This was thanks to the amazing work by the NVIDIA & Interact teams! In comparison, AMD vLLM did not work on day 0 for DeepSeekv4.1 Flash.Source: SemiAnalysis InferenceX AMD’s vLLM documentation points to using vllm/vllm-openai-rocm:deepseekv41-flash-0909, but from hour 0 of the model release to hour 23, AMD has not publicly released the image. AMD claims, “SPEED IS THE MOAT,” yet it has still not released it by the 23rd hour. We wish that, going forward, the AMD team has a better process for hour 0 model releases. Eventually when they did publicly release for “day 0” image support, performance-wise, it is currently up to 14.8x worse perf per dollar than H200 and up to 42x worse perf per dollar than B200/B300. The power of the CUDA MOAT is NVIDIA’s collaboration with its massive 6 million-developer community ecosystem including most of the vLLM & SGLang & Tokenspeed maintainers which means that CUDA is optimized on day 0. Overall, AMD did make significant improvements but the performance per dollar is still currently 2-4x worse than B200. AgentX Engram DRAM Offloading Improving Performance The HBM and DRAM offload use the same GPU kernel to select and dequantize rows. With HBM, it reads GPU memory; with Unified Virtual Addressing (UVA), it reads pinned host memory directly. Both support full decode graphs. Moving the table into HBM accelerates only the sparse lookup, leaving decoder computation and communication unchanged, resulting in little overall benefit while consuming memory otherwise available to KV cache. Another benefit of Engram offloaded to DRAM which means you can reduce the communication overhead by using less HBM GPUs per replica. For example, when enabling Engram offloading on B300, we are able to switch from TP4 to now TP2 which improves the pareto curve by up to 1.6x. When iso-model quality, less HBM is required as the engrams could be offloaded to host DRAM. Thus HBM bandwidth matters way more than HBM capacity. For inference workloads where memory bandwidth matters the most, 4-hi HBM provides the best $/bandwidth and therefore lowest cost per token. If China continues to make more and more revolutionary model architecture innovations, soon it could potentially 0Hi HBM stacks. [ ](https://newsletter.semianalysis.com/p/long-live-the-short-king-why-4-hi) [ Long Live the Short King: Why 4-hi HBM Wins ](https://newsletter.semianalysis.com/p/long-live-the-short-king-why-4-hi) Myron Xie, Bryan Shan, and 3 others · Sep 13 [ Read full story ](https://newsletter.semianalysis.com/p/long-live-the-short-king-why-4-hi) Moreover, on B300 and week-0 stack, moving the Engram table back to HBM did not improve results and stay within run-to-run variance. This is a result of the work optimizing DRAM offload, such as async, and overlap. SSD Offloading On B200, we created a unoptimized vLLM fork and stored the Engram tables in memory-mapped files on local SSDs. File backing lets the OS reclaim table pages when other applications need RAM. Pages already cached in memory can be served without reading the SSD again. Furthermore, note that we were unable to turn on GDS The unoptimized SSD implementation changes how rows reach the GPU. It copies row IDs to the CPU, deduplicates them, gathers the requested rows into pinned buffers, copies those rows back to the GPU and dequantizes them. This work runs between segments of the GPU execution graph. Native UVA performs row selection and dequantization directly on the GPU, avoiding the CPU round trip. A warm filesystem cache removes physical SSD reads, but leaves the coordination, row gathering and transfers. This is why a file already cached in RAM can still perform worse than a pinned DRAM table. The comparison measures the whole serving path; it does not separate the time spent on each of these operations. B200 DRAM dominates both measured SSD serving curves in total tokens per dollar and P90 interactivity. Near 125 tokens/s/user, DRAM delivers 121 million total tokens per dollar versus 52 million for SSD. For production serving, SSD offloading is likely not worth the tradeoff. On the B200 configurations we measured, SSD offloading loses on both measures: every observed SSD point has a DRAM alternative that delivers higher P90 interactivity and more total tokens per dollar. The only points where Cheaper storage does not automatically produce a cheaper inference service. Moving Engram to SSD leaves the same four expensive GPUs and the rest of the server in place. Reclaiming RAM only creates an economic benefit if it enables a cheaper server configuration or additional useful capacity. The current unoptimized path provides neither benefit, and the filesystem cache still consumes RAM when table pages are resident. Next we will look at the mechanisms and specific implementation of ngrams from DeepSeekv4.1 Flash, LongCat, Qwen3.8 Flash Next.
Disney employees have an AI dashboard to track who's 'tokenmaxxing'
Disney staffers have an "AI Adoption Dashboard." One Claude user invoked the chatbot 460,000 times in nine days.
OpenClaw creator: User asked for a token refund for errors
OpenClaw creator Peter Steinberger shared a refund request he received for errors, including "fabricated data." He refunded $0, as the agent is free.
Indeed exec details why they'll never have an AI token leaderboard
"We track token use, but we use it in the background," Indeed's chief information officer, Anthony Moisant, told Business Insider.
Small Business Tech News: Tokenmaxxing Wanes, Fixing Your “Spotty” Wi-Fi, Yelp Adds An AI Phone Assistant
“Tokenmaxxing” wanes in popularity, some non-profits are using AI in smart ways and Yelp adds an AI phone assistant
AI token leaderboards are 'dystopian': Replit AI chief
As Amazon and others rethink AI-use rankings, Replit's AI head said it was a poor measure of employee performance.
OpenAI reveals how much its researchers are spending on AI coding
OpenAI researchers spend heavily on AI coding agents, with top users burning through $7,000 daily in tokens.
Perplexity partners with Nvidia to launch Portable Computer, a fully local AI agent with zero token costs
Perplexity launches Portable Computer, a local AI agent platform that runs privately on Nvidia DGX Spark and RTX-powered Linux machines without cloud fees.
Datavault AI Provides Q1 2026 Business Update Highlighting Tokenization Adoption and Infrastructure Progress
More than $800 million in tokenization contracts signed; generating nearly $100 million in fees expected to be recognized in 2026 First Quarter…...
Are AI tokens the new signing bonus or just a cost of doing business?
Maybe tokens really will become the fourth pillar of engineering compensation. But engineers might want to hold the line before embracing this as a straightforward win.
Dell Deskside Agentic AI cuts token costs by up to 87%
Dell, with Nvidia, introduces Deskside Agentic AI to manage token costs, enabling local trillion-parameter model runs and reducing cloud expenses.
AI Driving Game Demo Already Putting Developers Into Debt
Hoping to launch next year, Teach My Little Sister How To Drive is prematurely sending its creators into the red over exhaustive token use
Is AMC Entertainment Holdings (AMC) Undervalued Following Its Robinhood Token Dispute?
AMC Entertainment Holdings (AMC) is back in the spotlight after a public clash with Robinhood over tokenized products tied to AMC shares and the possibility of regulatory scrutiny around those digital instruments. Against this backdrop, AMC Entertainment Holdings has been volatile, with a 1-day share price return of 4.24% and a 7-day share price return that declined 7.17%, while the 30-day move fell 2.77% and the 90-day share price return sits at 5.13%. The year-to-date share price return of...