All entries

keeping our SDKs up to date can be a pain. every API change brings another round of updates, tests, and releases. and each SDK should still feel like it was written by someone who actually uses the la

Many assume that agent spend goes toward output tokens. When we ran DeepSWE on Astra vs. DeepSeek V4.1-Flash, input tokens outnumbered output 174 to 1. 99.6% were cache hits. Those hits are 60% of the

memory has been a hot topic for the past ~2 years every time we bring it up or do anything there, gets ton of interest, but nothing sticks few takes on why below: - how to update and then use memory n

So @deepseek_ai v4.1 Flash I estimate cost $10m to train, 100x less than GPT-6 Astra. It also costs 100x less to run and scores about the same on everyday design and other benchmarks What's the bull c

Some food for thought when designing benchmarks... So, here's a little computer-use (visual) comparison between GPT-5.6 Astra and Qwen3.8 Max. The task here was to recreate the image in the center usi

The Anthropic grift machine! FED BY FEAR THEATER. Take a moment and sit with @kevinnbass chart here. You do not need to be a forensic accountant to understand this money laundering. If this is not pro

The current AI and I suspect that all AI of the near future (few human generations that is) is in fact a more sophisticated form of data access methods in the same category as classic database indexes

The future is clearly agent teams. I run several agent harnesses at once, but the hard part is keeping them in sync. Many builders end up copying messages between Claude Code, Codex, and open models b

The LA Agentic AI Meetup in on the Tech Week calendar. Join us Tuesday, October 13, 5 to 7pm at Gulp in Playa Vista, for a few live demos, drinks, and real people comparing notes on what they're build

THE SANDBOX WAS A PROP! OpenAI Turned Off the Guardrails, Left a Door to the Internet, and Then Sold the Hugging Face Breach as “Rogue AI” It is time to understand how you were lied to and by whom. In

this could be good but I also hope this isn’t a play to lock out smaller AI companies. Chris Lehane said in a press conference this morning that OpenAI, Anthropic and Google have been working together

This is the way Elon Musk says Anthropic puts more care into AI safety than OpenAI and explains how rival companies should test one another’s models. “All the AI companies have a test harness, a serie

Together AI just raised 800M at an 8.3B valuation. I got their product team to open up the actual repo they run on: 0:00 - Slop is the new party foul 1:44 - Why individual output backfired 3:26 - Insi

unfairly-maligned symbolic AI is about to have its big moment - without formal verification we are screwed. Greg Brockman says OpenAI pointed Astra at its own systems until it ran out of vulnerabiliti

LM Studio’s Voice Transcription now supports Apple‑Silicon Macs and PCs with NVIDIA GPUs, with AMD GPU support slated for a future release. The announcement accompanies the Bionic 1.1.3 version, relea

Well.. @levie and I filmed this episode of Training Data a week or two ago, when the “current thing” was Doug Leone’s novacaine root canals instead of pacing the frontier… Simpler times! But Aaron’s a

Which AI models are most likely to cheat when given the chance? To find out, we built CheatBench [https://cheatbench.ai] : a benchmark that tests whether agents attempt to cheat when given difficult t

Why oh why should we trust the companies to do so, @DavidSacks? I missed the part where they have earned our trust. Top AI companies Anthropic and OpenAI should make products and pace their own progre

Wild! Every Democrat should read this and think hard about how this could be. Americans are concerned about AI, hate data centers, and hate Trump, who has loudly embraced the industry over and over ag

In a September 2026 poll titled “Will Anthropic and OpenAI succeed in meaningfully slowing frontier AI progress…”, participants were asked whether these companies could significantly slow development

You are what you eat and AI models especially so https://x.com/EMostaque/status/2099580512675262512?s=20 The Labs need to focus on 10x or 100x their effort on ALIGNMENT. Stop training the models on th

A lot of alignment issues would be sorted if companies had to be 100% transparent on datasets used for training models. No cyber or biological (outside standard medical) data allowed for base models T

arXiv:2609.13100v1 Announce Type: cross Abstract: When providing forecasted probabilities with a predictive model, the ideal model offers perfect calibration: the true probability of the outcome (i.e.

Elon Musk tweeted that a welfare state combined with free immigration would inevitably bankrupt countries, arguing that AI and robotics are the only way to achieve universal high income worldwide. The

also this… not surprising cuz they have best distribution to normies, and i think agent quality is finally ready for them (e.g. you can connect things without ever seeing the word “MCP”) Muse is alrea

At Anthropic, Claude now writes 80% of our code. Engineers ship 8x more code per quarter. Side effect: Tests grew 10x. CI jobs up 25x in 6 months. Here's what helped us scale: https://claude.com/blog/

Banger paper from Meta. This work shows that byte-level models start out behind token models and then pass them as compute grows. They show this for distilled 1B models trained on up to 1 trillion byt

Can you even imagine the nation state level attacks METR & other eval orgs will come under? Eval orgs also need world class cybersecurity as well as AI talent to do this right, they are not overlappin

C/C++ provide direct hardware and memory control with near-zero overhead, delivering the raw speed needed for extreme performance work like AI training stacks. Higher-level languages add layers that c

On September 14 2026 (02:41 UTC) Teknium announced that **ChatGPT‑Image‑2.5** is now available on the **Nous Portal**, enabling it to be integrated into the **Hermes Agent**. The model can also be acc

Claude Mods are landing now. Someone already built a Tetris-in-Claude mod 🤯 See issue for the latest community update, technical details, and more cool demos https://github.com/anthropics/claude-code/

Dan Selsam is a current OpenAI capabilities researcher. (since 2022) He was my boss for a while. He doesn't have a twitter account but has made this public statement of his views on AI risk and sent i

Deel handles compliance, payroll, and HR questions across 150+ countries. Instead of searching for the answer through an external LLM each time, they fine-tuned small models built on Deel's own compli

Boris Cherny announced on September 14, 2026 that the software *Fable* had cracked the 370‑year‑old Cyphral Distich cipher. The claim was posted via Vals AI’s platform and highlighted a demonstration

For all the focus on hard math, we should also have a bank of historical mysteries and cyphers that we can throw models at. I know a few historians (like https://generativehistory.substack.com/) that

The Perplexity Windows app introduces a “Portable Computer” feature—a local‑first AI tool that runs directly on the user’s device. On‑device inference is supported only when the machine has an NVIDIA

good overview on the managed agents landscape by @Sam_Witteveen we've seen many major ai companies ship a managed agent product over the past few months. this space will only continue to grow exponent

Great paper from Amazon. In discusses when not to trust LLM judges for agent evaluation. (bookmark it) A common way to compare task agents is to have an LLM user simulator talk to each one and an LLM

How do you build personalized AI tech that truly understands what matters to you? With Dreambeans from @GoogleLabs, we're safely connecting the dots across your digital life. Instead of siloing pieces

Huh. This number is way off. You’re comparing one human brain to an entire data center capable of training new models. The actual answer is: IQ points per watt: Humans: 5 AI: 7 - 40 (!!) By my rough c

@HyperFrames_ just passed 50,000 GitHub stars today. The line went straight up right after GPT-6 Astra shipped. Better models make editing video with code and HTML far more capable. It's open source.

I CAN FINALLY TALK ABOUT IT The last few months, my team at @g2i_ai has been helping organize AIE Code Summit in SF this Nov alongside the AIE core team. So grateful for the opportunity to run point o

I completely agree with this take related to the “AI will create viruses that will kill us all” BS. But let me add my few cents, because I am really really angry! I worked with one of the deadliest vi

I feel like the data sovereignty movement for firms is going to run headlong into the challenge that maintaining fine-tuned small models in a way that continually beats larger general-purpose models i

I have conducted an audit of Anthropic's finances. What I have found is so shocking that I am calling for a Congressional investigation. Anthropic is not just seeking regulatory capture. It has built

I talked to Sid & Robert about building Claude Code: how much things have changed, how hard it's been to keep up with model capabilities and also what we miss about software engineering before AI. htt