Ankush

@benbbaldwin Garbage fad

I still do not understand the purpose of OpenClaw. Am I a boomer

These are the Twitter/X accounts with the most engagement so far in 2026. I suppose I had some intuition for how bad it was, but jeez, this is what you get when the ecosystem is broken.

@kylegoon Corrupt as balls

They started the war in Iran, which is breaking down because of the U.S. dependency on oil. Meanwhile, they’re spending a billion in taxpayer funds that will increase our dependency on fossil fuels. These guys don’t even pretend to have a coherent plan. The dumbest people.

@nypost You mean Conan the Destroyer

Arnold Schwarzenegger, 78, undergoing brutal gym prep for 'Conan the Barbarian' sequel trib.al/KSSi4Aq

@natashajaques @marwaabdulhai @isadorcw @yanming_wan @jzl86 @maxhkw Amazing work, thanks!

This is joint work with @marwaabdulhai @isadorcw @yanming_wan @jzl86 and @maxhkw. Project page: sites.google.com/view/llmwritin… Paper: arxiv.org/abs/2603.18161

The paper I’ve been most obsessed with lately is finally out: nbcnews.com/tech/tech-news…! Check out this beautiful plot: it shows how much LLMs distort human writing when making edits, compared to how humans would revise the same content. We take a dataset of human-written essays from 2021, before the release of ChatGPT. We compare how people revise draft v1 -> v2 given expert feedback, with how an LLM revises the same v1 given the same feedback. This enables a counterfactual comparison: how much does the LLM alter the essay compared to what the human was originally intending to write? We find LLMs consistently induce massive distortions, even changing the actual meaning and conclusions argued for.

Socrates in 460 BC: “The unexamined life is not worth living.” Marcus Aurelius in 150 CE: “You have power over your mind—not outside events.” Augustine of Hippo in 400 CE: “Do not go outside; return into yourself. In the inward man dwells the truth.” Marc Andreessen in 2026:

I look forward to future generations watching this clip in order to get some sense of how proudly stupid the self-proclaimed Great Men of our time actually were

Seriously impressive from a Tsinghua + Galbot research team

These numbers are pretty crazy: China exported over 1.5 million vehicles in Jan-Feb, which corresponds to ~9 million on an annualized basis. I checked, Japan at its peak exported 6.7 million vehicles a year (the 1985 record: ceicdata.com/en/japan/vehic…) and Germany never reached more than 4.4 million (in 2016: ceicdata.com/en/germany/mot…). Which means that China right now is not only the world's largest car exporter, it's the largest car exporter in history, and by a huge margin. And it happened super fast: just 5-6 years ago they were only exporting 1 million vehicles a year, now they do 1.5 million in just 2 months!

WSJ: To help make the case on Iran, Graham traveled several times to Israel in recent weeks, meeting with members of the country’s intelligence agency. “They’ll tell me things our own government won’t tell me,” he said. He spoke with Israeli Prime Minister Benjamin Netanyahu, coaching him on how to lobby the president for action. wsj.com/politics/polic…

We live in a country where an alcoholic from Fox News runs our military, a former heroin addict is in charge of our health, and they both report to a convicted felon. What a time to be alive.

@TheStalwart it's weird that people who found themselves going from non-coders to super-coders in one day think that they have some sort of "lead" on people who haven't opened the app yet

PSA (obvious, but just in case): no one uses Black-Litterman. Don’t waste time on this. Don’t teach this (I am working on a course for Columbia and will work to remove it). Don’t do complicated things.

@faustocoppi60 Bloody gorgeous

The spring is here, March 1st! 🥰 I hope you had a nice 10°C+ ride like I'm having at the moment.

It is hard to communicate how much programming has changed due to AI in the last 2 months: not gradually and over time in the "progress as usual" way, but specifically this last December. There are a number of asterisks but imo coding agents basically didn’t work before December and basically work since - the models have significantly higher quality, long-term coherence and tenacity and they can power through large and long tasks, well past enough that it is extremely disruptive to the default programming workflow. Just to give an example, over the weekend I was building a local video analysis dashboard for the cameras of my home so I wrote: “Here is the local IP and username/password of my DGX Spark. Log in, set up ssh keys, set up vLLM, download and bench Qwen3-VL, set up a server endpoint to inference videos, a basic web ui dashboard, test everything, set it up with systemd, record memory notes for yourself and write up a markdown report for me”. The agent went off for ~30 minutes, ran into multiple issues, researched solutions online, resolved them one by one, wrote the code, tested it, debugged it, set up the services, and came back with the report and it was just done. I didn’t touch anything. All of this could easily have been a weekend project just 3 months ago but today it’s something you kick off and forget about for 30 minutes. As a result, programming is becoming unrecognizable. You’re not typing computer code into an editor like the way things were since computers were invented, that era is over. You're spinning up AI agents, giving them tasks *in English* and managing and reviewing their work in parallel. The biggest prize is in figuring out how you can keep ascending the layers of abstraction to set up long-running orchestrator Claws with all of the right tools, memory and instructions that productively manage multiple parallel Code instances for you. The leverage achievable via top tier "agentic engineering" feels very high right now. It’s not perfect, it needs high-level direction, judgement, taste, oversight, iteration and hints and ideas. It works a lot better in some scenarios than others (e.g. especially for tasks that are well-specified and where you can verify/test functionality). The key is to build intuition to decompose the task just right to hand off the parts that work and help out around the edges. But imo, this is nowhere near "business as usual" time in software.

I'm honestly astonished that Citadel Securities published a rebuttal to the @Citrini7 piece. citadelsecurities.com/news-and-insig…

We’ve identified industrial-scale distillation attacks on our models by DeepSeek, Moonshot AI, and MiniMax. These labs created over 24,000 fraudulent accounts and generated over 16 million exchanges with Claude, extracting its capabilities to train and improve their own models.

@systematicls Good for you

renaissance technologies: 300 PhDs, secret algorithm, 30 years of data us: claude and a python library the gap is closing faster than they'd like

@TheStalwart Exactly.

I do loathe all the stuff about how you have to use AI or you're going to be left behind. If it's going to be that disruptive, then there's probably not much that you can do. And also there's no skill involved in using it. There's no learning curve. Can always pick it up later.

The stock market is up so we don’t have to prosecute pedophiles is one of the worst political messages i have ever heard.