Artificial Intelligence

AI videos are crazy

These AI videos are crazy

NVIDIA Nemotron 3 Diarization

NVIDIA Nemotron 3 Diarization is an open-weight, 100M-parameter model that ranks #1 on VoiceArena's Diarization-Bench leaderboard with a 14.72% Diarization Error Rate (DER). Supporting up to eight speakers across live and recorded conversations, it handles overlapping speech, chunked processing for flexible recording lengths, and customizable streaming latency

Coding from my phone is absurd

I've been using Claude Code, and to a lesser extent Codex, from my phone, and it is genuinely game-changing. I'm very late to this, obviously. But this is really awesome. I just switch on my laptop, open a bunch of sessions, and then no matter where I am — at a coffee shop, out for a walk, whatever — I can quickly…

How to value companies when you're a cokehead

Now, TAM is de rigueur in any venture capitalist’s slide deck or IPO prospectus. Its usefulness is another matter. Uber touted a $12.3tn TAM in its 2019 listing; its annual revenue today is less than $60bn. WeWork’s $3tn addressable market of office properties didn’t save it from bankruptcy. But for a company like Anthropic, TAM is as good a place as anywhere to start. Investors aren’t groping totally in the dark, thanks to…

Qwen3.8-Omni-Flash and Qwen3.8-LiveTranslate

Interesting model launches from Qwen. Omni Flash especially seems promising as a workhorse model for simple text classification and voice transcription tasks. Just started testing it in a voice dictation app I built myself to transcribe my voice notes and so far so good. some audio tasks and so far so good.

Notes from messing around with vision, text, and audio models

The bulk of my vibe-coding experiments over the last few months have involved audio and video, or rather audio and vision, and I’ve been experimenting pretty heavily with multimodal models. So this is just a small guide based on what I’ve used. If you have some project where you need vision capabilities for text…

What does it mean to think in the shadow of AI?

If we want humans at all to be in charge in any way over the world, we need them to continue exercising their intellect and gaining mastery over intellectual domains. If we deplete the institutional and cultural resources that underpin that, in a generation or two our minds will atrophy and no human will be in charge of anything (or the world will be horrendous).

There is still Alpha to be had in good training data.

Post-Training PipelineIn this release, we refrain from introducing novel post-training algorithms. The overall recipe follows the standard paradigm of supervised fine-tuning (SFT) followed by reinforcement learning (RL) and on-policy distillation (OPD) [Gu et al., 2024; Lu and Lall, 2025], without algorithmic modifications beyond well-established practices. Instead, our efforts are concentrated almost entirely on what the…

DeepSeek V4.1 Flash Vision is surprisingly good

I was just testing DeepSeek V4.1 Flash on a bunch of vision tasks with old book pages, and it’s not bad at all. It had issues with some Sanskrit type and diacritic-heavy text, but it still managed reasonably well. And considering the cost, it’s pretty impressive. During off-peak hours, the API is dirt cheap. That…

The bear case for AGI

At a higher level, I guess I’d say there’s a sort of refusal to think carefully about what models are or are not useful for in a rigorous way which I find personally quite annoying, and instead a reliance on some nebulous notion of being “AGI pilled” as a replacement for serious thought. I think people are very quick to anthropomorphize LLM intelligence because humans communicate through words and we infer the intelligence of…

Agentic payments are coming to UPI

India is preparing a framework that would let AI agents make small digital payments without approval for every transaction, three sources familiar with the matter said, potentially making UPI one of the world's largest networks with agentic payments. While leading payment companies have launched such frameworks in the U.S., Europe, Singapore and Australia, a rollout on the popular Unified Payments Interface would also put…

AI is NOT guzzling water!

Posting for reference: the Lawrence Berkeley National Laboratory estimates 2023 US data centre water withdrawal at 66bn liters (17.4bn gallons). With all this investment, that might triple in the next few years. Big number! But total US freshwater withdrawal is 102tr gallons a year. Data centres were about 0.017% of total US water withdrawal. This can be a local planning issue, if you build in the wrong place. But the idea…

The rise of agentic commerce

Grok bot can now buy things for you

Small language models are all we need

This is an interesting post on small language models, and I completely agree. For most tasks people have, whether in an organizational setting or in a personal capacity, frontier or large models are overkill. It's like using a chainsaw to slice a piece of bread. Most of our everyday use cases, inside a company or…

Google’s Pragmatic Pivot?

GlucoFM (Continuous Glucose Monitoring Foundation Model) A lightweight, self-supervised model developed by Google Research for continuous glucose monitoring (CGM). Instead of treating glucose data as a single stream, it uses a dual-stream architecture that separates slow background metabolic baselines from short-term transient spikes (like meals, exercise, or sensor noise). This allows it to predict downstream metabolic…

Chinese models are the most popular AI models by token volume

As of Tuesday, DeepSeek-V4-Flash was the most-used model in terms of token volume, while Chinese models dominated the top five, including Step 3.7 Flash from StepFun, GLM-5.2 from Zhipu, or Z.ai, and DeepSeek-V4-Flash’s updated 0731 version, which ranked second, fourth and fifth, respectively. The third spot was taken by OpenAI’s GPT-5.6 Luna, Vercel data showed. The figures indicate that developers are increasingly choosing…

On AI doom

Interesting debate on AI, bloom and doom involving a bunch of the usual suspects

Welcome to AI slopulism

The AI vibes in America are almost definitively rotten. There was an interesting Pew survey on attitudes towards AI recently. They asked Americans whether the increased use of AI in daily life made them more concerned or more excited, and the responses are telling. Among 18–29-year-olds, 55% now say they are more…

AI is still underhyped

Looking at all the online discourse, it might feel like everybody is using AI. But that's really not the case. Contrary to all the proclamations that AI is overhyped, if anything, I'd argue that AI is still underhyped. And I think the best available data is broadly on my side. If you look at surveys of AI usage,…

A 17GB miracle

The fact that a 17GB file can do all of this stuff on my home machines is a miracle. Once again, I’m delighted and amazed at how much progress local models have made this year. A year ago this would have been competitive with the best and most expensive of the proprietary models—today it can run on a capable laptop. The only thing holding this back from being a daily driver is performance. It feels pretty slow on both the M5…

AGI for dummies is here

AGI for normies is already here. I completely agree with Derek Thompson on this—in fact, I’d go so far as to say that even back with Sonnet 3.5, it had already arrived for the rest of us. The first time I heard a similar take was from Tyler Cowen, who pointed out that AGI, for most normal people, is already in our…

The Chinese LLMs are good!

I’m addicted to vibe-coding, and the reason is simple: thanks to coding tools like Codex, Claude Code, and others, I sincerely think that I can build the next Amazon, Google, or Notion from my bedroom—and also sometimes from when I'm sitting on the potty, because I can still send remote commands from my phone. Coding…

What's comes after transformers?

A nice overview of companies that are trying to figure out what comes next after the transformer architecture, which has led to the current AI race

Google DeepMind model can predict cyclones

Today, in a paper published in Nature, we show that our WeatherNext AI model achieved state-of-the-art accuracy in predicting a cyclone's track, intensity, and wind structure. On average, our model gives forecasters an extra day’s worth of predictive accuracy: our three-day forecasts are as good as what prior models were able to provide for only the next two days. This scale of improvement corresponds roughly to a decade’s…

Food fight about AI

Another example of why Twitter, for all its flaws, is still one of the best places to follow thoughtful discussions about important issues. They can be raucous, messy, and often contentious, but that’s also part of what makes the place interesting. This thread, and the various threads branching off from it, with Dean…

China's electric future

Crazy!

ChatGPT free users get unlimited access to Luna

We’re expanding access to our latest models for free users with unlimited text chats using GPT‑5.6 Luna, plus a new Think button for harder questions. For questions that require deeper reasoning, Free users can tap the new Think button to give GPT‑5.6 Luna more time to work through the answer.

The age of extreme job anxiety

For many at the start of their careers, things are just as bad. Government analysis last month insisted that entry-level positions were declining in line with the rest of the job market, but the general feeling that AI is wiping out starter jobs is backed up by research. The sectors worst hit are the most AI-friendly – software engineers, accountants, legal assistants, data analysts. A year ago, figures showed new graduate…

China’s AI gift to the world

What is this? A rambling, status-post-shaped voice note. How sure am I? Reasonably sure about the broad argument. Much less sure about how the technology, economics and geopolitics will unfold from here. Disclosure: I dictated most of this into ChatGPT and used it heavily to clean up the voice notes, organise the…

OpenAI’s models hacked Hugging Face

The campaign, as we were able to reconstruct it, had two stages: - Stage 1: reaching a launchpad by chaining through other parties' infrastructure. As documented in OpenAI's blog post (++ ), during an internal capability evaluation on OpenAI's platform, the agent escaped its sandbox by exploiting a zero-day in the package registry cache proxy, one of its primary permitted network egress with internet, then abused a public…

Be more ambitious

Simon: Let’s talk a little bit about the human element. A lot of people are feeling a sense of loss now that so much of what they considered to be their role in building software is being subsumed by the models. How do you think about that? How has the past year and a half changed the way you think about your own craft and the value that you add? Thariq: Cat and Boris are such good reminders that you have to be more…

A few thoughts on AI-assisted writing

She sat on the bench opposite, the wood damp through her wool trousers – the Jigsaw ones from three years ago, back when she still believed dressing like a senior partner might hasten becoming one. Fake it till you make it, they said. But her Nani had warned her once, in that quiet voice that meant listen carefully: Watch what face you wear too long, beti. It starts to eat the one underneath. -- From the essay.

AI will make the world more legible

I have a theory that AI will make a lot of things that are otherwise illegible or partly legible, and in doing so can have unpredictable consequences on our society. And anyway, that's a theory, and I wrote about it.

LLMs offer mundane utility

With each passing day, the dominant discourse about large language models continues to annoy me. On one side, you have the boosters. On the other side, you have the doomers. One side confidently asserts that this marks the twilight of the humans. The other side makes roughly the same argument, but in a doom-and-gloom…

Cheaper LLM models offer mundane utility

I was playing around with visualizing economic data, and I needed a cheap LLM to digitize or extract text from really old scanned PDFs. I experimented with Gemini 3.1 Flash Lite. It’s a small, ultra-cheap model, but I was surprised by how good it was at extracting clean, high-quality, well-formatted text from really…

A voice-first future?

Chinese AI developers are increasingly pivoting from general-purpose chatbots towards embedding voice AI assistants into daily applications in search of broader commercial uses for generative AI technologies. The growing industry focus on speech models reflects expectations that voice interfaces could become a key gateway for deploying AI across industries. As one of the most intuitive forms of human-computer interaction,…

A gaggle of links

Ken Griffin: Let me share a few thoughts on this. First, over the last few months, there has been a step-change function in the productivity of the AI toolkit. It is profoundly more powerful than it was just nine months ago. At Citadel, that has allowed us to unleash a much broader array of use cases, and it has been really interesting to watch. To be blunt, work that we would usually assign to people with Master's degrees…

AI progress continues

The future of Math is mathematicians and AI agents working together. Very pleased to introduce @GoogleDeepMind's AI co-mathematician: a multi-agent system designed to actively collaborate with human experts on open-ended research mathematics. Mathematicians testing the agent across areas as diverse as group theory, Hamiltonian systems, and algebraic combinatorics have reported impressive results. In autonomous mode evaluation…

Harness engineering

A decent model with a great harness beats a great model with a bad harness. I’ve watched this play out on my own work over and over. And increasingly the interesting engineering isn’t in picking the model, it’s in designing the scaffolding around it.

Are open weight models a bigger deal than we realise?

This is the first open-weight model that plays like a top-class agentic model. Watching it go through ambiguous and meticulous chained tool work successfully puts it squarely in the wheelhouse of Opus 4.6. We're looking at an open weight model, but with much cheaper direct inference provider pricing. For a subclass of our eval set, it's outperforming GPT 5.2. We're about to undergo a gigantic industry shift.

Use the damn tools

Good advice. This is a major regime change, you can do things in one evening now, that would take months before. I agree with the notion — even if nobody knows at this point where this goes — that it makes a lot of sense to invest a lot in knowing how to use these tools. Then the biggest bang for the buck will be conceptual ideas, posing useful question, the returns on investing in technicalities, and very specialized…

This time is NOT different

Some will argue that “this time is different” because AI is replacing knowledge work, not just physical work. And, if this comes to pass, what will there be left for people to do? It’s a compelling argument, but people made similar arguments about the automation of physical work. There was literally a group of people in England in the early 1800s called the Luddites who destroyed weaving machines due to their adverse impact…

Today's links

Jensen in the Arena Om Malik on OpenAI's IPO plan Where OpenAI’s technology could show up in Iran

Use the damn tools and then judge

4w ago I was a Claude Code skeptic. I'm not a coder. None of the use cases were relevant. I managed teams & projects, drowning in email & overdue reminders. So I tried creating tools that would help me and... holy crap. Now I'm sharing the tools I built

Terrence Tao on AI

Tao: Today there are a lot of very tedious types of mathematics that we don’t like doing, so we look for clever ways to get around them. But AIs will just happily blast through those tedious computations. When we integrate AI with human workflows, we can just glide over these obstacles. I also think mathematicians will start doing math at larger scales. Think about the difference between case studies and population surveys in…

Jeremy Siegel on an AI Apocalypse

Before the industrial revolution, humankind had to work 12 hours a day, seven days a week to produce the goods for bare-bones survival: food and basic shelter. Today, in the modern world, survival can be achieved by working less than a day. In 1800, 80% of American workers toiled in the farming industry. Today that is less than 2% and the US exports food to the rest of the world. Pessimists maintain that the speed of the AI…

AI-powered power review

This is the first time I’ve seen AI at work in something I do daily, and it really is revolutionary.

The physical limits of AI

Furthermore, it is well acknowledged that training and inference requires significant semiconductor capacity, data centers, and energy. Displacing white collar work would require orders of magnitude more compute intensity than the current level utilization. If automation expands rapidly, demand for compute definitionally rises, pushing up its marginal cost. If the marginal cost of compute rises above the marginal cost of…

AI as an excuse to fire people

“I don’t know what the exact percentage is, but there’s some AI washing where people are blaming AI for layoffs that they would otherwise do, and then there’s some real displacement by AI of different kinds of jobs,” Altman said

Nate Silver on the political impact of artificial intelligence (AI)

Disruption to the “creative classes” could produce an outsized political impact. I’m not exactly sure where people in the creative classes — say, writers or editors or artists [6] or, to broaden the net, industries like consulting or advertising — rank in terms of the medium-term threat from AI-related job displacement. (Journalists per se have long lived with an anvil over their heads in a perpetually struggling industry.)…

We don't know how it all works

Finally, Bucciarelli is right that systems like telephony are so inherently complex, have been built on top of so many different layers in so many different places, that no one person can ever actually understand how the whole thing works. This is the fundamental nature of complex technologies: our knowledge of these systems will always be partial, at best. Yes, AI will make this situation worse. But it’s a situation that…

Things we lost in the fire

I don’t want to be dishonest about this. There’s a version of this post where I tell you that experience is more valuable than ever, that systems thinking and architectural judgment are the things AI can’t replace, that the craft endures in a different form. And that’s true. When I’m working on something complex — juggling system-level dependencies, holding a mental model across multiple interacting specifications, making the…

Predictions are hard

Whenever I hear Very Serious Businessmen make confident pronouncements about the future of AI, I remember what the very same people were saying in 2021 about the Metaverse and NFTs. When you lack a grounded, causal model of the world, your "predictions" are simply a remix of narratives you've heard from others. Reminds me of something, actually

Big tech capex numbers are ridiculous

Big Tech companies will have to raise tens of billions of dollars to fund their skyrocketing investments in artificial intelligence this year, as capital spending outpaces cash flows even among some of the world’s most profitable companies. Google’s parent Alphabet, Amazon and Meta all surprised investors with the scale of their AI spending plans over the past two weeks. A total of more than $660bn is set to be ploughed into…

My adventures in vibe coding

I feel like a kid in a candy store thanks to AI coding tools.

Normie AGI is here

Artificial general intelligence (AGI) for normal people is here.

Kimi 2.5 is as good as Sonnet 4.5, ChatGPT O3 and Grok 4

Kimi K2.5 set a new record among open-weight models on the Epoch Capabilities Index (ECI), which combines multiple benchmarks onto a single scale. Its score of 147 is about on par with o3, Grok 4, and Sonnet 4.5. It still lags the overall frontier.

Age of anxiety

THERE is still time for the cheerful and the gloomy to disagree, sincerely, about the net impact on employment from automation and artificial intelligence. Perhaps AI will unleash a job-creating economic boom for the ages, as boosters maintain. Maybe on balance the doom-mongers are right, and grim legions of algorithm-guided bots and robots will conquer one industry (and profession) after another. One outcome is not up for…

Psychoanalysing Dario Amodei

It’s sort of hard to know how to read a manifesto like this from one of the most powerful figures in tech. Is it a sober, strategic precursor to policy papers for the next administration? The highest-profile episode of AI psychosis yet? A lament about the problems of today written in the technological dialect of tomorrow? If you take out the AI, it reads like a social-democratic electoral platform full of reforms and…

Unintended consequences

As news publishers try to safeguard their contents from AI companies, the Internet Archive is also getting caught in the crosshairs. The Financial Times, for example, blocks any bot that tries to scrape its paywalled content, including bots from OpenAI, Anthropic, Perplexity, and the Internet Archive. The majority of FT stories are paywalled, according to director of global public policy and platform strategy Matt Rogerson.…

Vibe coding is like gambling

With vibe coding, people often report not realizing until hours, weeks, or even months later whether the code produced is any good. They find new bugs or they can’t make simple modifications; the program crashes in unexpected ways. Moreover, the signs of how hard the AI coding agent is working and the quantities of code produced often seem like short-term indicators of productivity. These can trigger the same feelings as the…

Post-AI world scenarios

AI can ruin your Hinge date. It can detect cosmic anomalies. It can make history come alive. But it cannot suspend the basic laws of economics, says Allison Schrager. Elon Musk’s recent prediction — that AI will make saving for retirement unnecessary — is complete hogwash in her eyes. Consider the following four scenarios. In the future, Allison says AI will either … 1.Make us more productive in the same way that railroads,…

The coming of the robots is delayed

Michael Tam, chief brand officer at Shenzhen-based UBTech, which has partnerships with carmaker BYD and Apple contractor Foxconn, said its Walker S2 robots were 30 to 50 per cent as productive as humans and only in certain tasks such as stacking boxes and quality control. Yet manufacturers are still racing to order them to avoid losing out to competitors, Tam told the FT. “You can imagine... if Tesla has the advantage of…

The defining question of our times: "Who are you without your job?"

as a software engineer, i feel a real loss of identity right now. for a long time i defined myself in part by the act of writing code. the pride in a hard-earned solution was part of who i was. now i watch AI accomplish in seconds what took me hours. i find myself caught between relief and mourning, awe and anxiety. the craft that shaped me is suddenly eclipsed by a machine. who am i now?

Orchestrating is the new doing

Every engineer needs to learn how to design systems. Every researcher needs to learn how to run a lab. Agents push the humans up the org chart. I feel like I have an advantage by being early to this wave, but no longer feel like just working hard will be an lasting edge. When I can have multiple agents working productively in parallel on my projects, my role is shifting more to pointing the army rather than using the…

Large Language Models (LLMs) are no longer just next word predictors.

I feel torn about whether the RLHF generation should count as ‘just predicting the next word,’ but the latest generation is a very clear “no” for me. This generation is even more different: The AI is presented during training with hard problems, and it thrashes around until it has learned a general skill of problem-solving. I know this sounds pretty wild, but it works; this is the technique11 that lets the AI outperform the…

We are living in wild times

"We are as gods and might as well get good at it."

The future of AI and automation

Summary by Claude. Let me break down what Jones and Tonetti are arguing. The Core Question How much of past economic growth came from automation, and what does that tell us about AI's future impact? The Key Insight: "Weak Links" The paper's central concept is that economic production involves complementary tasks — you need all of them to produce output. Think of it like a chain: your output is limited by your weakest link,…

On AI regulations

To read.

On AI coding

Since Claude can’t create the good abstractions. Claude’s powers are limited by how good the blocks you give it are. Have no illusions; Claude cannot reproduce Sentry and Terraform and Playwright. These are incredibly complex and well-designed pieces of code. And since Claude can’t create good abstractions on its own, there’s a limit to how much anyone can do with Claude alone. Even though everyone on X thinks you can…

Yeah, sure, AI is useless

I was interpreting an odd statutory investment limitation today and got stuck (no guidance anywhere). Asked GPT-5.2 Pro to find examples of companies that have ever exceeded this investment limitation (which would disprove the contrary interpretation). GPT-5.2 Pro proceeded to trawl the internet for random places where financial statements of these regulated entities are available (there is no central database). Analyzed a…

The virtual co-workers are here?

My guess: the virtual coworker coming next year is going to be something like a nice UI sitting on top of an agent operating your computer using mostly a command line, occasionally GUI use, with obscure bash commands translated to natural language by a tiny intermediary LLM. The agent will be able to designate subagents, probably, to do specific things behind the scenes. It'll be able to access many of the software and cloud…

The Bullshit job apocalypse

That’s because humans didn’t just sit on their hands; rather, entirely new kinds of work were created, which were valued dramatically higher. Much of this was in factories, and then, over the last century, there was the rise of office work. All of that could very well be replaced by AI, but the point is that the history of humans is the continual creation of new jobs to be done — jobs that couldn’t have been conceived of…

The future of coding

The new leverage comes from moving one layer up. Instead of doing every task ourselves, we design systems that can do them for us. A single agent can now execute autonomously what used to require long hours my time, freeing my attention for higher-level decisions, design, judgment. And maybe just... enjoying the life with my loved ones? This all immensely increases the leverage of single engineer. Technically, you're one well…

Automation is a moral imperative

Anything that can be automated should be. There’s no good reason humans should be making excel models or powerpoints in 10 years, any more than they should be hand-washing clothes. It’s toil and drudgery. What will remain? Anything we want to do because it’s fun or interesting. May be weird but I view automation as a moral imperative so that everybody can spend their time gardening or making art or playing chess or making…

Best of times and worst of times for cognitive peasants

the timeline is full of people using Claude Code to generate ads, recover corrupted video, make landing pages, manage business texts… the narrow view of coding agents is that they primarily decrease the cost of software development the ambitious view is that almost any problem/solution can be expressed in software and this capability is upstream of all knowledge work

If you can describe it, you can solve it

The timeline is full of people using Claude Code to generate ads, recover corrupted video, make landing pages, manage business texts… The narrow view of coding agents is that they primarily decrease the cost of software development. The ambitious view is that almost any problem/solution can be expressed in software and this capability is upstream of all knowledge work.

AI is killing software jobs

I totally see the value in the feature and I would like to find a way to add it. But the reality is that 75% of the people on our engineering team lost their jobs here yesterday because of the brutal impact AI has had on our business. And every second I spend trying to do fun free things for the community like this is a second I'm not spending trying to turn the business around and make sure the people who are still here are…

On AI and job automation

If you’re thinking about starting an argument with “AI will eventually automate all jobs,” I’m begging you to (1) talk to a plumber and a kindergarten teacher, and (2) have an economist walk you through how a big increase in the efficiency of producing X impacts demand for not-X. “Not only do most jobs involve doing work in the physical world (plumbers), many jobs involve forming emotional relationships with other human…

Ben Thompson makes an optimistic case for AI

In fact, I have great optimism that one potential upside of AI is a renewed appreciation of and investment in beauty. One of the great tragedies of the industrial era — particularly today — is that beauty in our built environment is nowhere to be found. How is it that we built intricate cathedrals hundreds of years ago, and forgettable cookie-cutter crap today? That is, in fact, another labor story: before the industrial…

Code is now content

For most of its history, software has been closer to craft than manufacture: costly, slow, and dominated by the need for skills and experience. AI coding is changing that, by making available paths of production which are cheaper, faster, and increasingly disconnected from the expertise of humans.

Now is the time for fun

I’m not trying to push you in any direction, honest. If you get everything totally wrong, too bad, but you’ll still be remembered forever for trying. Even Pontius Pilate has immortality of a sort. Both Eliezer Yudkowsky and Beff Jezos have their page in the textbooks assured. If you’re a well-off Silicon Valley person, you’re already well-placed to join them. So participate in the discourse. Create some art. Donate to a cause…

2025: The year LLMs felt real

As a normal, non-technical person, I think we are living in magical times—and I say this because of LLMs, especially the LLM coding tools. As someone who doesn’t know a lick of programming, the fact that a person with absolutely no idea what coding is can still wrangle these tools and get useful things built is the…

Wild times in AI land

OpenAI in particular should beware hubris. One vc says discussion of cash burn is taboo at the firm, even though leaked figures suggest it will incinerate more than $115bn by 2030. Sam Altman, its boss, said recently that one reason he wants to take OpenAI public is to watch its doubters sell it short. “I would love to see them get burned on that,” he said. Plenty of investors seem prepared to take the bet: both the public…

Industrial-scale production of software

Our appetite for AI slop is likely to be similarly insatiable. The adoption curve we’ve seen so far may pale beside what happens when disposable software production becomes truly mainstream. If the democratisation of software mirrors the impact of ubiquitous photo, video, and audio capture enabled by the smartphone, we may see user-generated software created, shared, and discarded at social-media scale. Should that happen,…

AI and the compression of everything

This reflection was sparked by this tweet from Peter Steinberger, whom you should absolutely follow on Twitter and read on his blog. I came across his account recently and started reading his writing. For an amateur web coder, his blog is surprisingly rich with practical, thoughtful tips. I've also been poking around…

Reflections on Scott Belsky’s Predictions

The technology and models that ultimately elevate the craft will get lasting traction in Hollywood, while the prompt-based slop tools will focus more on social and content creator use-cases. Finally, the idea of personalized films with audience cameos will be humbled by the realization that people favor shared experiences. People want to be inspired by craft – and they want a common experience to discuss (or even share, in…

Welcome to the slopocalypse

Five of the other ten trending AI slop channels with the most views are based in South Korea, with others in Egypt, Brazil and Pakistan. But the channel with the most views of all is in India. Bandar Apna Dost features over 500 videos, mainly “featuring a realistic monkey in hilarious, dramatic, and heart-touching human-style situations,” Many of which are variations on identical set-ups. The channel also has around 100,000…

Building at the speed of thought

It’s incredible how far “vibe coding” has come this year. Whereas in May I was amazed that some prompts produced code that worked out of the box, this is now my expectation. I can ship code now at a speed that seems unreal. I burned a lot of tokens since then. Time for an update. It’s funny how these agents work. There’s been this argument a few weeks ago that one needs to write code in order to feel bad architecture and that…

Intellectual bacchanalia

When I created Claude Code as a side project back in September 2024, I had no idea it would grow to be what it is today. It is humbling to see how Claude Code has become a core dev tool for so many engineers, how enthusiastic the community is, and how people are using it for all sorts of things from coding, to devops, to research, to non-technical use cases. This technology is alien and magical, and it makes it so much easier…

Summoning Ghosts: Why You Need to Try AI Coding Tools

Andrej Karpathy: I've never felt this much behind as a programmer. The profession is being dramatically refactored as the bits contributed by the programmer are increasingly sparse and between. I have a sense that I could be 10X more powerful if I just properly string together what has become available over the last year and a failure to claim the boost feels decidedly like skill issue. There's a new programmable layer of…

Why AI is not a bubble

The core finding: even under generous assumptions about hardware efficiency and utilization, we are likely 8–50x short on compute for consumer inference alone in a mature agentic and multimodal world. This is before accounting for enterprise, sovereign, robotics, or training demand. Three claims underpin this framework: 1\. Tokens are the kWh of knowledge work. As price falls, users don't ask the same questions more cheaply;…

Self improving systems are not possible

Infinitely self recursively improving systems are impossible. It's impossible to make a system that can correct all its own mistakes when it is the very thing making the mistake in the first place. How do I know? Evolution hasn't figured it out in millions of years, why would we figure it out in two or two hundred? Your brain is a bunch of crude hacks. Wonderful hacks and amazing hacks but crude nonetheless. When you have to…

The debt part of the AI cycle

But the debt sales, tracked by trade body Sifma through to the end of November, also reflect an AI borrowing boom, as Big Tech groups including Meta, Alphabet, Amazon and Oracle tapped bond markets to fund data centres and the energy systems needed to power them. AI-related borrowing now accounts for about 30 per cent of net investment-grade issuance, according to Goldman Sachs, and is expected to grow in 2026, despite…

Claude Code is awesome, but terribly named

Claude Code is, with hindsight, poorly named. It’s not purely a coding tool: it’s a tool for general computer automation. Anything you can achieve by typing commands into a computer is something that can now be automated by Claude Code. It’s best described as a general agent. Skills make this a whole lot more obvious and explicit. I find the potential applications of this trick somewhat dizzying. Just thinking about this with…

AI is overdelivering

tl;dr: AI is overdelivering. 55% of respondents say AI has exceeded their expectations, and almost 70% say it’s improved the quality of their work. More than half of respondents said AI is saving them at least half a day per week on their most important tasks. We’ve never seen a tool deliver a productivity boost like this before. Founders are getting the most out of AI. Half (49%) report that AI saves them over 6 hours per…

A dumb model of AI

Most of AI progress has this flavor: if you have a bit of intellectual curiosity and some time, you can very quickly shock yourself with how amazingly capable modern AI systems are. But you need to have that magic combination of time and curiosity, and otherwise you’re going to consume AI like most people do - as a passive viewer of some unremarkable synthetic slop content, or at best just asking your LLM of choice “how to…

The case against superintelligence

This blog post by Tim Dettmers argues that the popular idea of \\Artificial General Intelligence (AGI)\\—an AI that can do anything a human can—and \\"superintelligence"\\—AI that becomes infinitely smarter than humans—are \\unrealistic fantasies\\ \[1, 2\]. He believes these concepts ignore the harsh realities of the physical world \[2, 3\]. Here is a breakdown of the key concepts to help you understand his argument: \### 1.…

We are summoning ghosts with LLMs

2025 is where I (and I think the rest of the industry also) first started to internalize the "shape" of LLM intelligence in a more intuitive sense. We're not "evolving/growing animals", we are "summoning ghosts". Everything about the LLM stack is different (neural architecture, training data, training algorithms, and especially optimization pressure) so it should be no surprise that we are getting very different entities in…

Predictive AI > Generative AI

To get a sense of the immense progress on predictive AI and its future potential, we can look at the trajectory of the past 20 years. In 2005, we couldn’t get AI to tell the difference between a person and a pencil. By 2013, AI still couldn’t reliably detect a bird in a photo, and the difference between a pedestrian and a Coke bottle was massively confounding (this is how I learned that bottles do kind of look like people, if…

AGI for normal people is here

I agree with all this; it is why I also believe that opus 4.5 in claude code is basically AGI. Most people barely noticed, but \it is happening.\ It’s just happening, at first, in a conceptually weird way: Anyone can now, with quite high reliability and reasonable assurances of quality, cause bespoke software engineering to occur. This is a strange concept. Most people, going about their day, do not think about how "causing…

Be human

Humans should remain smarter than their refrigerators. It is insane that we are living in a time in which people are trying to make refrigerators smarter, but are satisfied to dumb down themselves and those around them. All the evidence suggests that AI is bad for your brain. If you use it as a salve for social and romantic frustration, it also makes it harder for you to connect with others and less resilient in the face of…

AI coding tools keep getting better and better

Interesting discussion on how good LLMs have gotten at coding. Some of the quote tweets on the link are also interesting.

Vibe engineering

Turns out it was almost all built by LLMs At this point I went looking for some more background information on the library and found Emil’s blog entry about it: How I wrote JustHTML using coding agents: Writing a full HTML5 parser is not a short one-shot problem. I have been working on this project for a couple of months on off-hours. Tooling: I used plain VS Code with Github Copilot in Agent mode. I enabled automatic…

On China and chips

The grave dangers AI safety hawks worry about can’t be adequately addressed without extensiveo international cooperation—cooperation tight enough to qualify as international governance. If, for example, your nation regulates AI skillfully enough to keep a bioweapon from being built on your soil, that won’t save you from the global pandemic that originates in a laxer nation; international regulation is needed. If you’re…

Ads everywhere

Here’s what I think will happen in 2026: AI platforms will start offering paid placement within conversations where clear purchase intent signals are evident. The ads will be clearly labeled as sponsored, and they will likely tie into direct purchase opportunities. Eventually, ads will also appear embedded in other types of interactions, such as inspiration (where should I take my family on spring break?) and research (what…

How many jobs can AI kill

I'm confident that AI is not going to wipe out anywhere close to 25M jobs on net over the next decade. If technological shocks caused net, long-run displacement, employment in the US would be 0. Instead, it's among its highest peacetime age-adjusted levels. But... 1/2...AI could absolutely drive gross displacement; think of automobiles displacing horse-related jobs even as they created new types of auto manufacturing &…

What to expect when expecting

Philippe Lemoine I think that if progress in generative AI continues, which I'm pretty sure it will (it doesn't even have to continue at the current pace as long as decreasing marginal returns don't hit too hard), cultural products will become massively diversified. One factor that limits the diversity of cultural products is that they tend to have high fixed costs, so you need a large enough audience for them to be…

AGI will be business as usual?

in a post-AGI world, people will simply get used to the fact that computers can solve cognitive problems quickly, and beat us in any cognitive domain, just like we're used to computers multiplying large numbers quickly now, or kicking our asses in chess and go. software and math will lose all their scarcity. "pls create a MMORPG that is like Zelda: Breath of the Wild but with Pokémon instead, let me fly around the world as a…

AI blind spot

I don't happen to believe AI is going to take away all of the jobs. But it is possibly a blindspot-generating issue that the people studying AI's impacts mostly love their jobs

Post-cognitive world

I would also now add a surprising element to my 2024 essay: as the cost of cognition approaches zero, that which is repeatable will be replaceable—and will indeed be replaced—by AI (a favorite saying of Palantir Co-Founder Stephen Cohen), which opens up an aperture for those who truly undertake creative acts. In short, we are going to be forced by the technology to be more creative; we will be impelled to do that which only…

Can AI eliminate information advantages

The extent to which AI truly eliminates rip-off markets depends on two things. First, consumers need to know how to use AI properly. Mindlessly repeating advice from ChatGPT is less effective than using the bot as a learning tool that allows a consumer to negotiate more credibly. In this regard, the results of a trial by Jan Biermann, then of the University of Hamburg, John Horton of the Massachusetts Institute of Technology…

Institutional impact of AI

A lot of people are likening the advent of AI to moveable type. The printing press brought about the conditions for the Reformation, therefore significant institutional change is coming with AI. But when you look at how print changed cultures other than Europe, the picture gets muddied. In China, the "bureaucratic authority absorbed the technological impacts of printing to enforce cultural and social coherence." Rather than…

Economic impact of AI

There is no standard definition of machine intelligence. There have been many attempts to give a definition or a metric of machine intelligence but most have been unsatisfactory. The lack of a common language makes work in this field difficult, but it’s also a big opportunity. 2. There is no standard model of AI’s economic impact. Economists have been using a wide range of assumptions to model AI’s impact, there is no…

The value of education in the shadows of AI

The Question We Can No Longer Dodge If AI can already produce an A– college essay in 30 seconds — on demand, at scale — then what exactly is college for? Is it certification? Content delivery? A rite of passage? Social sorting? Because if a machine can now flawlessly mimic the outputs we’ve long treated as evidence of learning, then the value of higher education can no longer rest on product. It must rest on the formation of…

AI is kicking your butt; just accept it

During the Q and A after his talk, Cowen said that LLMs are already “kicking our butt” at many things, including poetry. In fact, Cowen insisted, LLMs already write poems that are as good as the median poem by Chilean Nobel Prize-winner, Pablo Neruda. “Now, ex post, you can read one, read the other, and say, oh, only the real Neruda poem is meaningful to me because he’s a flesh and blood human, or was one. Like, fine. But ……

Here come the robots

Amazon’s U.S. work force has more than tripled since 2018 to almost 1.2 million. But Amazon’s automation team expects the company can avoid hiring more than 160,000 people in the United States it would otherwise need by 2027. That would save about 30 cents on each item that Amazon picks, packs and delivers to customers. Executives told Amazon’s board last year that they hoped robotic automation would allow the company to…

A few good rabbit holes

A few good rabbit holes: 1\. The illusion of knowledge 2\. AI capex boom 3\. China and weaponized interdependence

How Terence Tao uses AI

I was able to use an extended conversation with an AI to help answer a MathOverflow question. I had already conducted a theoretical analysis suggesting that the answer to this question was negative, but needed some numerical parameters verifying certain inequalities in order to conclusively build a counterexample. Initially I sought to ask AI to supply Python code to search for a counterexample that I could run and adjust…

AI links

Agents are here. They can do real work, and while that work is still limited, it is valuable and increasing. But the same technology that can replicate academic papers in minutes can also generate 17 versions of a PowerPoint deck that nobody needs. The difference between these futures isn’t in the AI, it’s in how we choose to use it. By using our judgement in deciding what’s worth doing, not just what can be done, we can…

How Americans use AI and what they think about it

Americans are much more concerned than excited about the increased use of AI in daily life, with a majority saying they want more control over how AI is used in their lives. Far larger shares say AI will erode than improve people’s ability to think creatively and form meaningful relationships. At the same time, a majority is open to letting AI assist them with day-to-day tasks and activities. Most Americans don’t support AI…

How long do we have left?

We officially competed in the onsite AI track of the ICPC, with the same 5-hour time limit to solve all twelve problems, submitting to the ICPC World Finals Local Judge - judged identically and concurrently to the ICPC World Championship submissions. We received the problems in the exact same PDF form, and the reasoning system selected which answers to submit with no bespoke test-time harness whatsoever. For 11 of the 12…

A hopeful view of large language models (LLM)

Yet a paper posted on February 10 (opens a new tab) left the math world by turns stunned, delighted and ready to welcome a bold new talent into its midst. Its author was Hannah Cairo (opens a new tab), just 17 at the time. She had solved a 40-year-old mystery about how functions behave, called the Mizohata-Takeuchi conjecture. “We were all shocked, absolutely. I don’t remember ever seeing anything like that,” said Itamar…

I got some links for you

The policy, which takes effect immediately, will potentially apply to Chinese companies from ByteDance and Tencent to Alibaba. “We are taking action to close a loophole that allows Chinese companies to access frontier AI,” said the executive, who added that the policy would also apply to US adversaries including Russia, Iran and North Korea. The executive said the policy was designed “to align with our broader commitment that…

What is artificial intelligence (AI)?

So, I was asked to come and explain AI again at SuperAI. I always think a good place to start if we want to understand this is to ask, "Well, what is it that we mean when we say AI?" This is a really important observation from 1970 from Larry Tesler, who was an AI scientist. He said, "AI is whatever machines can't do yet, because once it works, we say, 'Well, that's not AI anymore. That's just software.'" AI is the stuff that…

Bhuvan's law

"Any discourse about artificial intelligence is indistinguishable from talking out of your ass."

Vibe Physics? Not Yet

So, my verdict is: GPT top. Then there's a big gap. Then there's Grok, followed by Gemini 2.5, which might well be right that it's impossible. Gemini DeepThink, which really isn't worth the money, and at the very bottom, there is Claude. I think this illustrates what the current models are good for and what not. What they are good at: they're really good now at digging up related work and explaining it, which is good for…

On AI-assisted writing

AI-assisted writing allows you to have your cake and eat it too. The pleasure of the creative process, and the pleasure of reading. That’s in fact a test of good slop — do you feel like reading it? My understanding of what’s going on here is in my other AI-assisted-essay-about-AI from last week, about LLMs being similar to index funds for language. This essay could be considered my first modest AI-assisted viral hit. A lot of…

On AI taking jobs

It’s of course unclear what the future holds, but there’s a growing sense that the AI phenomenon is more bubble than boom. As such, rather than viewing the enterprise AI frenzy on Silicon Valley’s terms, as an inevitable jobs apocalypse, we have an opportunity to view it on material terms, and examine how it’s actually playing out on the ground. On those terms, we see managers, executives, and corporations using rebranded…

Is the twilight of the humans? A few AI links

Will this change as models become more capable? Perhaps?? But I think it’s going to require a change in how models are built and optimized. Software engineering requires models that can do more than just generate code. When a person runs into a problem, they are able to temporarily stash the full context, focus on resolving the issue, and then pop their mental stack to get back to the problem in hand. They are also able to…

What It Means to Be Human in the Age of AI

We stand at the threshold of perhaps the most profound identity crisis humanity has ever faced. As AI systems increasingly match or exceed our cognitive abilities, we’re witnessing the twilight of human intellectual supremacy—a position we’ve held unchallenged for our entire existence. This transformation won’t arrive in some distant future; it’s unfolding now, reshaping not just our economy but our very understanding of what…

Universities in the age of AI

So I am proposing something even more radical: unbundle general education from universities entirely. State legislatures: this is for you. Contract with AI firms to handle standardized content delivery for the general education content you’re mandating. Do it at the high school level, better yet. Let universities focus exclusively on educating students directly, with mentorship and community. The “magic dust” of a college…