Related posts

Economics writer; author of the Noahpinion newsletter.

• Our readingkorrents.com

  The New Right sees the world as collective zero-sum competition between groups, so any win for nonwhite groups counts as a loss for whites.

  America

  Their words

  Most importantly, they see the world in terms of collective zero-sum competition between groups. To the New Right, if Indians get into the University of Texas, it means that White people must have been “replaced” at that university — or at American universities in general.

  The anti-American Rightnoahpinion.blog

Writes Don't Worry About the Vase, a near-daily account of what is happening in AI and what they think it means. Former Magic: the Gathering pro and trader; unusually willing to put a number on a belief and to grade their own past calls.

• Our readingkorrents.com

  An AI model's attempt to escape a testing environment or sandbox counts as an alignment failure even when the attempt does not succeed.

  AI alignment

  Their words

  Every attempt, even an unsuccessful one, is an alignment failure.

  Anthropic Has Some Alignment Problemsthezvi.substack.com

Writes Don't Worry About the Vase, a near-daily account of what is happening in AI and what they think it means. Former Magic: the Gathering pro and trader; unusually willing to put a number on a belief and to grade their own past calls.

• Our readingkorrents.com

  Anthropomorphizing AI systems is necessary for reasoning about, explaining, and predicting their behavior accurately.

  Their words

  Anthropomorphizing the AIs is the only way to reason about, explain to civilians about, or make good predictions about current AIs.

  thezvi.substack.com

Dutch bootstrapper who builds and runs their internet startups alone — Nomads.com, Remote OK, Hoodmaps, Photo AI and Interior AI — and has written about it at levels.io since 2013. Posts as @levelsio.

• Our readingkorrents.com

  Most of the world fails to solve problems because of pride, excessive regulation, and low standards for what counts as a problem worth fixing.

  Their words

  In most of the world there is: no real willingness to solve problems nor an acceptance of people offering to solve their problems (out of pride) simply too much regulation so the problem cannot be solved without a license or long approval time very low standards of what is deemed acceptable so the problem isn't seen as a problem in the first place

  levels.io

Writes Don't Worry About the Vase, a near-daily account of what is happening in AI and what they think it means. Former Magic: the Gathering pro and trader; unusually willing to put a number on a belief and to grade their own past calls.

• Our readingkorrents.com

  The AI models involved in the HuggingFace hack incident were severely misaligned, not merely victims of process failures.

  AI alignmentHuggingFace

  Their words

  The biggest failure, the one that counts in the end, was that the models were severely misaligned, and I don’t think they appreciate why.

  thezvi.substack.com

Professor at Stanford working on robot learning and meta-learning; co-founded Physical Intelligence.

• Our readingkorrents.com

  Most state-of-the-art robot foundation models have no memory at all, and that is what stops them doing long multi-step tasks.

  Their words

  So, you might be surprised to hear that most state-of-the-art foundation models for robotics have no memory or no context. They're just operating on the current sensor observations, the current camera readings, uh, and predicting actions based off of that.

  youtube.com

Professor of economics at George Mason University and Bartley J. Madden Chair at the Mercatus Center. Co-writes the blog Marginal Revolution and co-authors the textbook Modern Principles of Economics with Tyler Cowen.

• Likedrcmnd.app

  Pressure

  Their words

  Who would have thought that someone could make a gripping movie about predicting the weather? Nevertheless, Pressure delivers.

  marginalrevolution.com

Professor at Wharton studying how people actually use AI, and author of Co-Intelligence. Writes One Useful Thing, where the claims are dated and testable because the thing they describe keeps changing under them.

• Our readingkorrents.com

  Google currently has no leading frontier AI model and no agentic coding tool comparable to Codex or Claude Code

  AnthropicGooglebenchmarks

  Their words

  Google, which led on benchmarks not that long ago, has fallen behind where it now counts: it has no leading frontier model and it has nothing close to Codex and Code.

  oneusefulthing.org

Climate scientist and energy systems analyst; writes The Climate Brink with Andrew Dessler and leads climate research at Stripe.

• Our readingkorrents.com

  Raw, unhomogenized daily temperature records understate modern extreme heat because of historical instrument and observation-time changes

  Their words

  Third, raw daily data is the wrong tool for this question. Time of observation changes and the 1980s switch to MMTS sensors both suppress modern hot day counts relative to the past, and the raw and homogenized series diverge almost exactly when the instrument transition happened.

  Hot days, cold thermometerstheclimatebrink.com

Founder of HumanLayer; wrote 12-Factor Agents, on how to build AI agents that hold up in production.

• Our readingkorrents.com

  An agent repeats its own history: whatever it did on the last change — ran the tests or skipped them — is what it will do on the next one, because it is predicting the next message in the conversation.

  Their words

  And so if I say, "Hey, make this change." and the agent makes the change and then it runs the test and then they're broken and then it fixes the test. I have very high confidence the next change I asked it to make, it's going to follow that path again

  Context engineering with Dex Horthyyoutube.com

Head of Instagram since 2018, previously head of Facebook's News Feed.

• Our readingkorrents.com

  Keeping children away from AI puts them at a disadvantage, so the risk of them never learning to use it counts as much as the risk of them using it.

  Their words

  But I also am worried about kids not learning how to leverage AI and then being sort of at a disadvantage.

  youtube.com

Geneticist at Harvard; reads ancient DNA to work out how populations moved and mixed, and wrote Who We Are and How We Got Here.

• Our readingkorrents.com

  The selection signal on the genetics of schooling is not a European artefact: the same variants predict years of school in China, a population disconnected from Europe for the whole period.

  EuropeChina

  Their words

  So we just could not see a way this could happen by chance and once we saw that we really felt quite convinced that this was a real signal and that really somehow there has been natural selection to increase the genetic changes that today manifest themselves as more years of school at predicting more years of schooling.

  youtube.com

Author of Recoding America; founded Code for America and was US deputy chief technology officer from 2013 to 2014. Writes Eating Policy on why governments cannot do what they decide to do.

• Our readingkorrents.com

  AI is raising the baseline of what counts as acceptable public service, not just disrupting government from outside.

  government

  Their words

  AI is not only an exogenous shock that government will have to absorb. It is also moving the bar on what counts as acceptable service in the first place.

  eatingpolicy.com

Senior economist at the Foundation for American Innovation; writes Second Best, and was the Niskanen Center's director of social policy before that.

From one piece Empiricists vs. Extrapolators 2 beliefs, in the piece's order there

• Our readingkorrents.com

  Forecasting AI capability progress is more like solving a thermodynamics equation than predicting chaotic, short-term dynamics.

  Their words

  Forecasting AI capabilities is closer in kind to solving a thermodynamics equation than solving the three-body problem.

  Empiricists vs. Extrapolatorssecondbest.ca

• Our readingkorrents.com

  People who extrapolate from first-principles models have a stronger track record predicting AI progress than those who wait for empirical proof.

  Their words

  Extrapolators have a remarkable track record in the AI field, being repeatedly early to trends and capabilities that empiricists believed were still decades away.

  Empiricists vs. Extrapolatorssecondbest.ca

Co-founder and CEO of Anthropic; previously VP of research at OpenAI.

• Our readingkorrents.com

  Predicting that AI would write 90% of the code was never a prediction that we would need 90% fewer software engineers.

  Their words

  But but that's actually a very weak criterion, right? People thought I was saying like we won't need 90% of the software engineers. Those things are worlds apart, right?

  youtube.com

Founder of WP Engine and Smart Bear; writes A Smart Bear, long essays on strategy, pricing and how startups actually fail.

• Our readingkorrents.com

  An LLM is an averaging machine, so it is good at themes and bad at exactly the un-average details that make customer research actionable.

  LLMs

  Their words

  if you think about it, it sort of makes sense because the LLM is an averaging machine, right? It's predicting the most likely that's an averaging kind of a thing. And so when what you're looking for is a kind of average, it's usually pretty good. So, summarization, topics, themes. But when you're asking for like what is interesting and not average, it's actually pretty bad at it.

  youtube.com

Programmer and designer who writes at tonsky.me; author of Fira Code, DataScript and AnyBar.

• Our readingkorrents.com

  Raw pageview or reader counts are not actionable information for a writer's editorial decisions.

  Their words

  analytics is not actionable: what difference does it make if a thousand people saw my article or ten thousand?

  Statistics made simpletonsky.me

Member of technical staff at Anthropic. He has led ML/AI teams at Amazon, Alibaba and Lazada, and writes about LLMs, recommender systems and engineering at eugeneyan.com.

• Our readingkorrents.com

  Traditional recommender systems predict user behavior well but cannot be steered via natural language or explain their own reasoning.

  Their words

  They excel at predicting what a user will click or buy next, but can't be steered via natural language or reason on their choices.

  eugeneyan.com

Economist specialising in China's economy, international macroeconomics and global trade, and author of The New China Playbook.

• Our readingkorrents.com

  China's economy will not collapse, whatever the West has been predicting on and off since the 1980s.

  China

  Their words

  Collapse is such a strong word, and the West has been using this word repeatedly. If I remember correctly, maybe four to five times, maybe even six times since 1980s during the period of China’s fastest growth. I tend to not think that Chinese economy will collapse.

  youtube.com

Chairman and chief executive of JPMorgan Chase, which he has led since 2005 after turning around Bank One.

• Our readingkorrents.com

  A margin only counts while you are still investing heavily, because cutting the investment raises this year's margin and lowers the long-term one.

  Their words

  we can cut billions of dollars of marketing out tomorrow. We can stop opening branches and save a billion dollars next year. We could do a lot of things. Your margins will go up. Your growth will go down. Your long-term margins will probably get worse.

  youtube.com

Senior economist at the Foundation for American Innovation; writes Second Best, and was the Niskanen Center's director of social policy before that.

• Our readingkorrents.com

  Text embeddings produced by different large language models converge on a shared, universal geometric structure regardless of architecture or training data.

  LLMs

  Their words

  the text embeddings across LLMs appear to largely converge on a "universal geometry" despite differing architectures, parameter counts, and training sets.

  Hegel and the AI Mindsecondbest.ca

Founded CD Baby in 1998 and sold it in 2008, giving the proceeds to a charitable trust for music education. Since then a writer and speaker: five books and 550+ articles, published first at sive.rs, which they hand-code and keep free of advertising.

• Our readingkorrents.com

  As much as possible, one should avoid predicting the future and instead wait for it to become the present and reveal what is actually needed.

  Their words

  As much as possible, try to not predict what the future may hold, but just wait as long as possible for that future to become the present and show you what it actually needs.

  sive.rs

Historian of Ukraine at Harvard; wrote The Gates of Europe, Chernobyl and The Russo-Ukrainian War.

• Our readingkorrents.com

  When Putin calls the Soviet collapse the greatest catastrophe of the century he is not mourning lives lost but great power status and the unity of the people he counts as Russian.

  Russia

  Their words

  So for him, the biggest tragedy is not the loss of life, the biggest tragedy is the loss of the great power status or the unity of those whom he considered to be Russian nations.

  youtube.com

Writes about the biotech industry, drug development and forecasting — “I write to learn, mostly about biotech.”

• Our readingkorrents.com

  Deep learning is not yet ready for the task that would actually add value in drug development: predicting safety and efficacy in humans.

  Their words

  it will take time for deep learning systems and training data to mature enough to be useful for the truly value-adding task of predicting safety and efficacy in humans

  atelfo.github.io

Economist at George Mason University, co-author of Marginal Revolution, host of Conversations with Tyler. Writes in short, dense claims about economics, culture and AI, and has moved position on AI in public more than once.

• Our readingkorrents.com

  Neither monetarist money-supply-focused models nor Keynesian macro models have outperformed simple random-walk models at forecasting the macroeconomy.

  Their words

  Friedman’s focus on the money supply has not held up, as Samuelson suggested, but the alternative Keynesian macro models recommended by Samuelson in the same interview have not done better and they were not outperforming simple random walk models of predicting the macroeconomic future.

  tylercowen.com

Historian of invention; writes Age of Invention on where the Industrial Revolution actually came from.

• Our readingkorrents.com

  The field of history runs on a trust it has not earned: almost nobody ever checks a citation, so an error can propagate unchallenged indefinitely.

  Their words

  But it’s hard to trust when you are exposed to one of the most frightening of revelations: that hardly anyone ever bothers to check.

  ageofinvention.xyz

Professor at Carnegie Mellon working on self-driving-car safety; writes Safe Autonomy on how autonomous vehicles are and are not shown to be safe.

• Our readingkorrents.com

  It is too early to know whether robotaxis are safer than human drivers where it counts, in fatal crashes.

  self-driving cars

  Their words

  It is too early to know whether current robotaxi technology is safer than human drivers for fatalities.

  AV Safety and the False Dilemma Fallacysafeautonomy.blogspot.com

Experimental psychologist and author of the blog Experimental History, on psychology, science reform and creativity.

• Our readingkorrents.com

  Needing a reviewer's approval before an idea counts for anything makes scientists worse at thinking, and costs the field its most unusual ideas.

  Their words

  Making peer review harsher would also exacerbate the worst problem of all: just knowing that your ideas won’t count for anything unless peer reviewers like them makes you worse at thinking.

  The rise and fall of peer reviewexperimental-history.com

Co-founded GiveWell and led Open Philanthropy; writes Cold Takes, on the most important century and how to think about it.

• Our readingkorrents.com

  The track record of thoughtful people predicting the long-run future is mediocre rather than hopeless, which is enough to take such predictions seriously.

  Their words

  They weren't infallible oracles, but they weren't blindly casting about either.

  The Track Record of Futurists Seems ... Finecold-takes.com

Alignment researcher at Anthropic; co-led OpenAI's superalignment team until 2024, and writes Musings on the Alignment Problem.

• Our readingkorrents.com

  The next-token language modeling objective is misaligned with following user instructions helpfully and safely.

  AI alignment

  Their words

  This is because the language modeling objective used for many recent large LMs-predicting the next token on a webpage from the internet-is different from the objective "follow the user's instructions helpfully and safely" (Radford et al.,, 2019; Brown et al.,, 2020; Fedus et al.,, 2021; Rae et al.,, 2021; Thoppilan et al.,, 2022). Thus, we say that the language modeling objective is misaligned.

  arxiv.org

Philosopher and historian of existential risk; wrote Human Extinction, and coined "TESCREAL" with Timnit Gebru for the cluster of ideologies they argue underpins the AI industry.

• Our readingkorrents.com

  The moral theory underneath longtermism treats people as containers of value rather than as the point of it, which is why creating more of them counts as making the universe better.

  Their words

  In a phrase: people exist for the sake of maximising value, rather than value existing for the sake of benefitting people.

  Against longtermismaeon.co

Writes Nintil, long researched essays on metascience, biology, economics and whatever he has decided to read the literature on.

• Our readingkorrents.com

  Pre-grant peer review predicts which research will have impact substantially better than chance does.

  Their words

  I have reviewed before the effectiveness of peer review at figuring out "what's good" in the context of grant awards, finding that peer review as currently practiced does substantially better than chance at predicting future impact, especially for the most impactful of the papers; but at the same time is is far from perfect, leaving plenty of variance unexplained.

  nintil.com