Scott analyzes whether AI forecasters will plateau near human levels or keep improving, arguing that even non-miraculous progress could yield 4-12 percentage point improvements in prediction market accuracy. Longer summary

Scott examines whether AI forecasters will plateau near current human performance or continue improving far beyond it. He analyzes a 2010 study by Daniel Reeves showing prediction markets only marginally outperformed simple statistical models, but argues this doesn't necessarily limit future progress. The post reframes those results to show they actually represent meaningful gains in difficult-to-predict domains like sports, then applies this reasoning to geopolitical forecasting. Using three different 'anchors' - comparing AI improvements to past technological gains, human-to-superhuman forecaster improvements, and chess AI progress - Scott estimates that maximally capable AI forecasters could improve prediction market accuracy by 4-12 percentage points over current levels, which he argues would be genuinely valuable even if not miraculous. Shorter summary

Scott's monthly links roundup covering AI progress, the coming flood of philanthropic funding to effective altruism from AI companies, new research on persuasion and forecasting, and various scientific and cultural topics ranging from dementia prevention to Romanian politicians' embarrassing quotes. Longer summary

This is the second part of Scott Alexander's monthly links roundup for July 2026, covering a wide range of topics including AI developments, effective altruism funding waves, healthcare research, and cultural curiosities. Major themes include Chinese AI models catching up to the US (though still 6-12 months behind), the coming "third wave" of American philanthropy from AI company equity donations potentially flooding effective altruism with $40 billion/year, AI systems now outperforming humans at persuasion in specific contexts, and various scientific findings from dementia prevention to infinite ethics. The post maintains Scott's characteristic style of jumping between serious technical analysis, historical oddities, and humorous observations, with extensive linking to sources and ongoing debates in the rationalist and effective altruist communities. Shorter summary

Scott examines the rise of AI superforecasters that now match or exceed top human forecasters, explores how they work and their current performance, and analyzes implications for decision-making, prediction markets, and the future role of AI opinions. Longer summary

Scott discusses the emergence of AI superforecasters that are now matching or slightly exceeding top human forecasters in prediction accuracy. He describes how these AI systems work (using scaffolding around frontier models like GPT/Claude), demonstrates their use with examples from FutureSearch and Preseen, and analyzes their performance on platforms like Metaculus where they're competing in tournaments against humans. The post explores both near-term implications (AI forecasters being easier to access than human superforecasters, potentially influencing policy and business decisions) and longer-term possibilities (AI forecasters serving as an 'opinion layer' for AI systems, transformation of prediction markets into AI-vs-AI competitions). Scott argues these developments could be genuinely beneficial, giving people access to superhuman forecasting ability just as AI threatens other aspects of society. Shorter summary

Scott examines the Metaculus Threat To Democracy Index, which uses crowdsourced forecasting to measure democratic health, discussing its advantages over traditional expert-based indices and its vulnerabilities to manipulation and selection bias. Longer summary

Scott discusses the Metaculus Threat To Democracy Index, a new approach to measuring American democratic health using crowdsourced forecasting rather than expert opinion. The index aggregates 153 questions about democratic threats (like election cancellations or ballot access) weighted by forecasters' historical accuracy. Scott analyzes its advantages—transparency, resistance to ideological bias, and ability to measure probabilities rather than binary outcomes—alongside its risks, including susceptibility to crowd attacks, bot manipulation, and question selection bias. He finds the recent data somewhat reassuring (showing no expected worsening during Trump's term) and suggests the index will become more valuable over time as question biases become less relevant. Shorter summary

Scott reviews the state of prediction markets after explosive growth, finding that most volume is degenerate sports gambling rather than useful forecasting, and proposes both technical solutions and two potential futures for the field. Longer summary

Scott Alexander examines the current state of prediction markets after their recent explosion in popularity, noting that while volume has grown from millions to billions per month, most of it comes from sports betting rather than the epistemic improvement he'd hoped for. He explores several problems: markets aren't asking the most important questions society needs answered; resolution criteria disputes ("rulescucking") create controversy; and there are concerns about insider trading and manipulation. He proposes solutions including a novel approach to conditional markets and suggests two paths forward: either creating user-generated, subjectively-resolved real-money markets (the "Siskind Cube"), or accepting that prediction markets' main value may be as training data for AI forecasters that could make the markets themselves obsolete by late 2026. Shorter summary

The 2026 ACX/Metaculus prediction contest is now open with a $10,000 prize pool, featuring reader-submitted questions, and Scott announces the winners of the best question submission contest. Longer summary

Scott announces the 2026 ACX/Metaculus prediction contest is live, with questions suggested by ACX readers covering US politics, AI, international affairs, and culture. Predictions must be submitted by January 17th to be eligible for the $10,000 prize pool, though participants can continue updating forecasts afterward for site leaderboards. The post also lists the winners of the best question submission contest, with prizes ranging from $150 to $700. Shorter summary

Scott asks readers to suggest questions for the 2026 ACX/Metaculus forecasting contest, with prizes for the top ten contributors and AI bots competing this year. Longer summary

Scott announces the upcoming ACX/Metaculus forecasting contest for 2026 and calls for question suggestions from readers. The contest has been running for several years with Metaculus handling most of the organization, and last year attracted over 4500 forecasters predicting on 33 questions. Scott explains that good questions should be objective, verifiable outcomes that will be resolved by year's end, providing examples of well-specified versus poorly-specified questions. The top ten question contributors will receive prizes ranging from $150 to $700, and this year's contest will include AI bots competing alongside humans. Shorter summary

Scott Alexander presents his monthly collection of interesting links and developments across technology, science, culture, and policy, with personal commentary on each item. Longer summary

Scott Alexander shares a collection of 61 interesting links, spanning topics from AI development and scientific studies to cultural observations and policy discussions. The post covers diverse subjects including GPT-5's reception, genetic research, religious demographics, urban development, mental health studies, and various scientific discoveries. Scott provides commentary and analysis on many of these items, often connecting them to broader themes or offering his perspective on controversial issues. Shorter summary

Scott provides a detailed analysis of OpenAI's attempt to convert from nonprofit to for-profit status, including the legal challenges, competing offers, and implications for AI development. Longer summary

The post explains the complex situation around OpenAI's attempt to convert from a nonprofit to a for-profit structure. Scott details the history of OpenAI as a nonprofit, why they want to change, and the various legal and financial challenges they face. The post covers Sam Altman's proposed buyout plan, Elon Musk's competing offer and lawsuit, and the role of state Attorneys General in approving any conversion. The post also explains the implications for AI development and safety, and contrasts OpenAI's structure with Anthropic's different approach to balancing profit with beneficial AI development. Shorter summary

Scott shares 50 diverse links covering topics from AI developments to policy changes, with commentary on current events, research findings, and societal trends. Longer summary

In this monthly links roundup, Scott Alexander covers a wide range of topics including AI developments, public policy changes, scientific research, and current events. The post follows a numbered format presenting each item with Scott's characteristic analysis and often humorous commentary. Notable topics include AI safety developments, changes in government policy, climate change updates, and various social and technological trends. The tone alternates between analytical and personal reflection, particularly when discussing potentially concerning political developments. Shorter summary

The 2025 ACX/Metaculus Forecasting Contest is now open, featuring 36 questions and a particular focus on comparing forecasting bots with human predictors. Longer summary

Scott Alexander announces the opening of the 2025 ACX/Metaculus Forecasting Contest, although the 2024 results are pending due to complications. He mentions hoping to compare Metaculus with Polymarket but couldn't due to recent events including FBI raids. Scott expresses particular interest in seeing how new forecasting bots will perform against top human forecasters in this year's contest, which includes 36 questions and offers $10,000 in prizes. Shorter summary

Scott examines the likelihood and potential severity of an H5N1 bird flu pandemic, analyzing prediction markets and historical data to estimate a 5% chance of pandemic in the next year with most likely moderate severity. Longer summary

Scott Alexander provides a comprehensive overview of H5N1 bird flu and its pandemic potential. He starts by explaining what flu is and its history of pandemics, then focuses on H5N1's current situation, analyzing prediction market estimates for its chances of causing a pandemic. The post examines different mortality scenarios and their likelihood, using data from past flu pandemics and current cases. It concludes with specific predictions about H5N1's future impact, suggesting a 5% chance of human pandemic in the next year, with varying degrees of severity if it occurs. Shorter summary

Scott Alexander praises Polymarket's election success but argues their Trump odds were mispriced, explaining why Trump's win doesn't significantly validate their numbers over other forecasters. Longer summary

Scott Alexander congratulates Polymarket for their success during the recent election, but argues that their Trump shares were mispriced by about ten cents. He uses Bayes' Theorem to explain why Trump's victory doesn't significantly vindicate Polymarket's numbers. Scott compares the situation to non-money forecasters like Metaculus versus real-money markets like Polymarket, explaining why he initially trusted the former more. He discusses the impact of a large bettor named Theo on Polymarket's odds and addresses several objections to his argument. Scott concludes that while prediction markets are valuable, they can sometimes fail and require critical thinking. Shorter summary

The post examines the spectacle of US elections, analyzes recent developments in prediction markets, and discusses various election-related forecasts and their implications. Longer summary

This post discusses the intense atmosphere surrounding Election Day in the United States, comparing it to historical spectacles and highlighting the emotional and psychological impact on the population. It then delves into prediction markets and forecasting, particularly focusing on recent events in Polymarket where a large bet by a single individual caused significant market movements. The post also covers legal developments regarding prediction markets, discusses various election-related predictions, and concludes with a poetic reflection on Election Day. Shorter summary

Scott Alexander presents a diverse collection of 41 links and news items, covering topics from science and politics to history and technology, with brief commentaries and insights. Longer summary

This post is a collection of 41 diverse links and news items covering a wide range of topics. The items include scientific studies, political developments, historical anecdotes, technological advancements, and cultural phenomena. Scott Alexander presents these with brief commentaries, often adding his own insights or expressing skepticism about certain claims. The tone is informative and occasionally humorous, with Scott pointing out interesting connections or implications of the information presented. Shorter summary

Scott Alexander examines prediction markets suggesting that replacing Biden as the Democratic nominee could significantly improve the party's chances in the 2024 election. Longer summary

Scott Alexander analyzes prediction markets to assess whether replacing Joe Biden as the Democratic nominee would improve the party's chances in the 2024 election. He finds that replacing Biden with Kamala Harris would be neutral to slightly positive, while replacing him with Gavin Newsom or a generic Democrat could increase their odds of winning by 10-15 percentage points. Scott discusses potential objections to these findings, examines his own previous skepticism about Biden's cognitive decline, and reflects on the implications for Democratic party leadership and decision-making. Shorter summary

Scott Alexander reviews recent developments in prediction markets and forecasting, including regulatory changes, platform pivots, and debates about the field's future. Longer summary

Scott Alexander reviews recent developments in prediction markets and forecasting. He discusses the CFTC's move to further restrict prediction markets, Manifold Markets' pivot to a sweepstakes model, a superforecasting report on COVID-19 origins, and debates about the future and value of forecasting. The post also covers various prediction market probabilities on current events and links to other forecasting news. Shorter summary

Scott Alexander analyzes the results of his 2023 forecasting contest, comparing various prediction methods and individual forecasters. Longer summary

Scott Alexander reviews the results of his 2023 annual forecasting contest, where participants predicted 50 questions about the upcoming year. He discusses the winners in both 'Blind Mode' (relying on personal knowledge) and 'Full Mode' (using aggregation algorithms). The post analyzes the performance of various forecasting methods, including individual forecasters, prediction markets, superforecasters, and aggregation techniques. Scott concludes that Metaculus, a forecasting platform, outperformed other methods, though some individual forecasters showed exceptional skill. He also examines which 2023 events were most surprising to forecasters and shares his main takeaways from the contest. Shorter summary

Scott Alexander analyzes the astronomical costs and resources needed for future AI models, sparked by Sam Altman's reported $7 trillion fundraising goal. Longer summary

Scott Alexander discusses Sam Altman's reported plan to raise $7 trillion for AI development. He breaks down the potential costs of future GPT models, explaining how each generation requires exponentially more computing power, energy, and training data. The post explores the challenges of scaling AI, including the need for vast amounts of computing power, energy infrastructure, and training data that may not exist yet. Scott also considers the implications for AI safety and OpenAI's stance on responsible AI development. Shorter summary

The post examines the performance of prediction markets in elections, current political forecasts, and various other prediction markets, while also discussing the challenges and potential of forecasting. Longer summary

This post discusses several topics related to prediction markets and forecasting. It starts by examining a claim that prediction markets have an 'election problem', showing that real-money markets performed poorly in recent elections. The author then analyzes current polls and prediction markets for the 2024 US presidential election, noting discrepancies between different platforms. The post also explores a forecasting experiment on AI futures, and reviews several other prediction markets on current events. Finally, it includes short links to other forecasting-related news and reflections. Shorter summary

The post discusses recent developments in prediction markets, including challenges in market design, updates to forecasting platforms, and current market predictions on various topics. Longer summary

This post covers several topics in prediction markets and forecasting. It starts by discussing the challenges of designing prediction markets for 'why' questions, using the OpenAI situation as an example. It then reviews the progress of Manifold's dating site, Manifold.love, after one month. The post also covers Metaculus' recent platform updates, including new scoring systems and leaderboards. Finally, it analyzes various current prediction markets, including geopolitical events, elections, and the TIME Person of the Year. Shorter summary

Scott Alexander defends effective altruism by highlighting its major accomplishments and arguing that its occasional missteps are outweighed by its positive impact on the world. Longer summary

Scott Alexander defends effective altruism (EA) against recent criticisms, highlighting its accomplishments in global health, animal welfare, AI safety, and other areas. He argues that EA has saved around 200,000 lives, equivalent to ending gun violence, curing AIDS, and preventing a 9/11-scale attack in the US. Scott contends that EA's achievements are often overlooked because they focus on less publicized causes, and that the movement's occasional missteps are minor compared to its positive impact. He emphasizes that EA is a coalition of people who care about logically analyzing important causes, whether broadly popular or not, and encourages readers to investigate and support the most beneficial causes. Shorter summary

Scott Alexander discusses recent developments in prediction markets, including regulatory issues, new platforms, and current market probabilities on various topics. Longer summary

This post covers several topics in the world of forecasting and prediction markets. It starts with an update on Kalshi's request to the CFTC for permission to create prediction markets about upcoming elections, discussing the public comments received. The post then announces the Manifest 2023 conference on forecasting and prediction markets. It discusses prediction markets related to a potential room temperature superconductor discovery, and how these markets are being used to track the developing story. The post also covers PredictIt's legal battle with the CFTC, the use of forecasting technology in finance, new forecasting platforms like Fatebook and The Base Rate Times, and several current prediction market probabilities on various topics. It concludes with some short notes on recent developments in the forecasting world. Shorter summary

Scott Alexander analyzes the surprisingly low existential risk estimates from a recent forecasting tournament, particularly for AI risk, and explains why he only partially updates his own higher estimates. Longer summary

Scott Alexander discusses the Existential Risk Persuasion Tournament (XPT), which aimed to estimate risks of global catastrophes using experts and superforecasters. The results showed unexpectedly low probabilities for existential risks, particularly for AI. Scott examines possible reasons for these results, including incentive structures, participant expertise, and timing of the study. He ultimately decides to partially update his own estimates, but not fully to the level suggested by the tournament, explaining his reasoning for maintaining some disagreement with the experts. Shorter summary

The post explores AI forecasting capabilities, compares prediction market performances, and provides updates on various ongoing predictions in technology and politics. Longer summary

This post discusses recent developments in AI forecasting and prediction markets. It covers a study testing GPT-2's ability to predict past events, reports on Metaculus' accuracy compared to low-information priors and Manifold Markets, and updates on various prediction markets including those related to AI development, abortion medication, and Elon Musk's role at Twitter. The author also mentions new features in prediction platforms and research on forecasting methodologies. Shorter summary