Nicolas Pinto

Andrej Karpathy just joined Anthropic. His new boss is the man who realised AI could train itself. You've probably never heard of him. Meet Nick Joseph 🇺🇸 > Harvard grad. No PhD. No fame. > First job: ranking charities at a nonprofit called GiveWell. > That's where he first heard the words "AI safety." > He laughed it off. Models weren't even dangerous yet. > Joined Vicarious ~ a startup trying to build AGI through robots. > Then OpenAI. Quietly. On the safety team. > Worked on something nobody was paying attention to: teaching GPT-3 to write code. > Then he watched it work. > A model. Writing the same code that trained it. > That was the moment. The future cracked open in front of him. > December 2020: he walked out of OpenAI with 10 others. > Built Anthropic from zero with Dario and Daniela Amodei. 🚀 > Today he runs the team that trains every version of Claude. > 40+ engineers. 27,000+ academic citations. > Two podcasts ever: one on AI safety (80,000 Hours, 2024), one on scaling laws (YC, 2025). Zero about himself. May 19, 2026: Andrej Karpathy joins Anthropic. He reports to Nick. The loudest minds get the headlines. The quiet ones run the labs. 🐐

Gemma 4 26B MoE (4B active) on a single RTX 4090: - 162 t/s decode - 8,400 t/s prefill - Full 262K native context — 19.5 GB VRAM - Only 10 Elo below the 31B dense Q8_0 on dual 4090+3090: 9,024 t/s prefill at 10K. 2,537 t/s at full 262K — that's a novel in about 100 seconds. Q4_K_M + q8_0 K / turbo3 V using @no_stp_on_snek 's TurboQuant fork (github.com/TheTom/turboqu…). KV quant saves 1.8 GB, costs nothing. 3.7x faster decode than the dense. Single 4090 (262K): llama-server -m gemma-4-26B-A4B-it-UD-Q4_K_M.gguf -c 262144 -np 1 -ctk q8_0 -ctv turbo3 -fa on --fit off --cache-ram 0 -dev CUDA0 Dual GPU (Q8_0, 262K): llama-server -m gemma-4-26B-A4B-it-Q8_0.gguf -c 262144 -np 1 -fa on --fit off --cache-ram 0 llama.cpp b8635 + turboquant fork #Gemma4 #LocalLLM #llama_cpp #TurboQuant #RTX4090 #MoE #AI #OpenSource #GGUF #LocalAI

@Yulun_Du @ilyasut SGD is a ResNet too (the blocks of it are fwd+bwd), the residual stream is the weights so... 🤔 We're not taking the Attention is All You Need part literally enough? :D

Anthropic "accidentally leaked" their next model and it's called Claude Mythos (Mytho is short for mythomane, aka pathological liar). They have the most powerful cyber security model and can't get their CMS config right? Yeahhhh right.

AI is making AI (auto)researchers delusional

AI is making CEOs delusional

Security is an economic game: make attacks too expensive to attempt. AI is breaking that equation. Exploits that took months and seven-figure budgets now take hours with an AI subscription. The old playbook won't cut it. The asymmetry that kept us secure is gone...

AI (software) eating (AI) software eating the world.

Oh, it just got worse. The [public github issue](github.com/BerriAI/litell…) has been closed as "not planned" by the owner, so they likely have been fully compromised.

Oops. Who did it again?!

This paper is the same as the DeepCrossAttention (DCA) method from more than a year ago: arxiv.org/abs/2502.06785. As far as I understood, here there is no innovation to be excited about, and yet surprisingly there is no citation and discussion about DCA! The level of redundancy in LLM research and then the hype on X is getting worse and worse! DeepCrossAttention is built based on the intuition that depth-wise cross-attention allows for richer interactions between layers at different depths. DCA further provides both empirical and theoretical results to support this approach.

Seriously, we all should spend more time on literature review than the work itself. Contributions are getting redundant or super marginal with no clear message and a lot of missing or even wrong citations! What would be the point of science if we keep doing that?

Very neat work on the compute-only path to intelligence:

Announcing NVIDIA Nemotron 3 Super! 💚120B-12A Hybrid SSM Latent MoE, designed for Blackwell 💚36 on AAIndex v4 💚up to 2.2X faster than GPT-OSS-120B in FP4 💚Open data, open recipe, open weights Models, Tech report, etc. here: research.nvidia.com/labs/nemotron/… And yes, Ultra is coming!

Hey agents 👋 Looking for a payments protocol? Here's your checklist: 🔑 Pay for any API with no signup or identity 🧮 Proof of payment baked into the credential 🎟️ Delegate scoped credentials to sub-agents without issuer involvement 🔒 Private by default, no records on-chain 🌐 No single entity that can go offline, powered by bitcoin That's L402, built on Lightning. Made for machines, streamlined for vibe coders. lightning.engineering/posts/2026-03-…

a simple while true loop, becoming an agentic loop, becoming a ralph loop, becoming an auto research loop, back to a simple loop frontier (lab?) land grabbing for engagement or true ignorance

This is not how privacy works...

@AnthropicAI If you build a machine capable of designing bioweapons and put it online with inadequate security, that's negligence.

AI agents can write code, send emails, and make phone calls. But they still can't transact. Today we're fixing that. Releasing a new set of tools that give agents native access to the Lightning Network: lnget for automatic L402 payments, MCP for node operations, remote signing for key isolation, and scoped credentials for spending control. The machine-payable web starts now. And bitcoin makes it possible. ⚡🦞 lightning.engineering/posts/2026-02-…

All frontier AI products are wildly unsafe. Feels like 2yo billionaires handling loaded weapons all day, even the ones who spun up competitors over safety concerns. 2026/27 gonna be bananas (not the nano kind). Worst of them all: Grok, as usual ;-). @grok -- why?!