foom!
foom! Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor. It’s happening faster than we t
Knowledge catalogue
foom! Our internal data shows Claude is accelerating AI development—a possible path to recursive self-improvement, or AI autonomously building a more capable successor. It’s happening faster than we t
Safety by narrow control has shown to fail many times. Need more transparency on the absolute frontier, and openness close behind. I found another API that offers claude-oceanus-v1-p the pricing and t
In early May, the best superforecasters predicted that, by the end of the year, the longest METR 80% task horizons would reach 3-4 hours. In late May, Claude Mythos achieved that number. We also asked
Anthropic's Claude 3.5 Sonnet (M3) model features sparse attention mechanisms, 1 million token context length, and multimodal capabilities, with Together AI optimizing the serving infrastructure to en
shower thought If: 1. AI is smarter than humans at law, therapy, etc. 2. Humans still like talking to other humans. Then: Humans are just an AI wrapper. Everyone should just regurgitate what Claude te
Anthropic PBC is expanding a program that enables organizations to test their cybersecurity defenses using its Claude Mythos Preview model. The initiative, which is known as Project Glasswing, launche
Big paper on AI coding agents using Github & other data The auto-complete tools (Copilot) led to 2.2x more code, local agents like original Claude Code led to 7.4x, & current remote coding agents 17.3
arXiv:2603.16572v2 Announce Type: replace-cross Abstract: Agent skills extend local AI agents, such as Claude Code and OpenClaw, with additional functionality. Their growing popularity has led to dedi
arXiv:2606.01490v1 Announce Type: cross Abstract: We present a controlled experiment evaluating 12 multi-agent LLM collaboration topologies for software architecture design. Using a 2imes2imes2 factor
arXiv:2606.00288v1 Announce Type: new Abstract: Large language models are undergoing a transition from model technology to system technology. As developers use Codex, Claude Code, AutoGPT, and related
new in deepagents: agent rubrics! you define a rubric, and the agent self-evaluates and iterates until it satisfies every rubric criterion. this is similar to /goal in claude code or codex, but more f
Prompting agents to 'do better' is unreliable 🙅 Giving them a rubric, a grader, and a correction loop is much closer to how you get your agent do what you want! Similar to /goal in Claude Code or othe
Five million users would agree. Resetting the limits tomorrow morning to celebrate. Time to go /fast nothing like switching to claude for a few days to try out a new model and going back to codex xhig
I’ve been saying this from day 1. Despite their launch video getting 50M views, Cowork was @AnthropicAI biggest mistake. I've been loyal to Claude but it would be really nice to have these tabs be con
The efficiency frontier! Where do you think GPT-5.6 will land? Claude Opus 4.8 has landed on DeepSWE Bench, posting a 58% Pass@1 and taking #2 overall behind GPT-5.5. It continues a broader trend: sli
The most interesting thing about Anthropic's 65B Series H announcement is this line (emphasis mine): Since our Series G in February, adoption has continued to grow across global enterprise customers,
Command A+ sets a new high for Cohere's machine translation capabilities. Opening a clear gap over open source peers Mistral Medium 3.5, DeepSeek, & OpenAI's gpt-oss, as well as Claude Opus 4.6. A+ al
arXiv:2605.28876v1 Announce Type: cross Abstract: CI failure logs are large (median 5k lines, max 200k in this corpus) and noisy. Coding agents that try to debug them depend on an upstream tool to red
models underestimate how much work it takes (token usage) to accomplish a task, just like us 🧵 Claude-Opus-4.8 takes you too much tokens - but is this issue general across agents? Do agents know how m
Nvidia up 0.7%, on news that tokenmaxxing is dead and H200 rental prices are down. What an absurd time to be alive. In the last 30 days alone: – Microsoft cancelled most of its Claude Code licenses, c
arXiv:2605.30353v1 Announce Type: new Abstract: Are AI agents tools, co-authors, or researchers? We present a quantified case study (N=1): a physicist supervising an AI coding agent (Claude Code, Sonn
arXiv:2605.29601v1 Announce Type: cross Abstract: As autonomous agents become more capable of performing real-world tasks, distinguishing scheming behavior from benign task pursuit may become a centra
When you are talking to an LLM, you are speaking to a synthesized work of interactive fiction, not a real being. ChatGPT, Claude, and Sydney are not their neural networks. If any LLM claims to be cons
arXiv:2605.29744v1 Announce Type: new Abstract: The impressive performance of generalist large language models (LLMs) such as GPT and Claude in healthcare raises a critical question: will domain-speci
Also out today: You can now directly configure the effort level and adaptive thinking in Code (/effort) and Cowork! Effort allows you to tune Claude's intelligence vs token spend, trading off capabili
Madison Mills / Axios: Anthropic says it expects Mythos-class models to be available to all customers “in the coming weeks” following the development of stronger safeguards — Anthropic released Claude
Big migrations and refactors are some of a team's most important work, and the easiest to push off to a 'better time' since they'd tie up engineers for a quarter. With dynamic workflows, Claude can no
glad to know Mythos' safety concerns have been addressed right as Anthropic also secured tens of billions in inference compute 👍 JUST IN: Anthropic announces it will roll out Claude Mythos “in the com
Vercel announced support for Claude Opus 4.8 on its AI Gateway, enabling developers to integrate Anthropic's latest large language model through Vercel's unified API platform. This update allows seaml
arXiv:2605.27439v1 Announce Type: cross Abstract: AI assistants like ChatGPT and Claude are recommendation engines, not search engines: they answer commercial queries by directly nominating brands rat
wait… if most people think 5.5 is better than 4.7, i assume that’s due to terminal coding benchmark… 4.8 is still outperformed by 5.5 Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper ju
Who's going first Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.
Someone just asked me “What I don’t get is why you see it as some kind of failure that AI labs are now using harnesses and neurosymbolic tools…?” I don’t! As I argued in my newsletter on Claude Code (
this basically destroys the extrapolations that had Anthropic making two trillion dollars a year. It's clear that growth for coding tools such as Claude Code has decelerated from the pace it was since
We have, as far as I can tell, no good tests of the productivity impact of the autonomous coding tools that appeared starting in December 2025. Every paper out there is from prior to the Claude Code/C
arXiv:2605.23772v1 Announce Type: new Abstract: Agentic systems have recently emerged as state-of-the-art approaches for automated theorem proving in formal mathematics. To assess how far these capabi
arXiv:2605.23904v1 Announce Type: new Abstract: Agent skills today are hand-crafted, generated one-shot, or evolved through loosely controlled self-revision, none of which behaves like a deep-learning
@GaryMarcus Can't agree more. OpenAI seems struggling to build any successful product after ChatGPT. Codex is promising but it's still early, and it's following Claude Code. It's questionable whether
this is even easier oh even better, since you can test the runtime without an LLM, you can post this into Claude: “look up http://activegraph.ai, install and run a small experiment I would like levera
In this post, we show you how OPLOG developed three AI agents using the Strands Agents SDK, deployed them to Amazon Bedrock AgentCore, and integrated Amazon Bedrock with Anthropic’s Claude Sonnet and
New inflection point in the accelerating growth of open-source models usage is coming 🦔Microsoft canceled its internal Claude Code licenses this week after token-based billing made the cost untenable,
Token Maxxing, April 2026-May 2026, RIP 🪦 🦔Microsoft canceled its internal Claude Code licenses this week after token-based billing made the cost untenable, even for a company with effectively infinit
arXiv:2605.20176v1 Announce Type: new Abstract: Large language models (LLMs) and agentic systems have shown promise for clinical decision support, but existing works largely assume that evidence has a
arXiv:2410.18856v4 Announce Type: replace Abstract: Frontier large language models (LLMs), such as GPT-5, Claude 4.5, Gemini 3, Llama 4, and DeepSeek-R1, represent a transformative class of AI tools c
for anyone curious, this was the result of many experiments bouncing around but this version uses... initial ideation: chatgpt for ideas/@replit for quick build final repo build: opus 4.7 prompting cl
.@trq212 is a builder's builder. After research at MIT, exiting his company, raising millions for another, and exploring research threads as an SPC member, he's now on the team building Claude Code. H
arXiv:2605.17554v1 Announce Type: new Abstract: Frontier deep research agents (DRAs) plan a research task, synthesize across documents, and return a structured deliverable on demand. They are being de
Today at Google I/O, Google released Gemini 3.5 Flash. This one skipped the -preview modifier and went straight to general availability, and Google appear to be using it for a whole lot of their key p
arXiv:2605.16286v1 Announce Type: cross Abstract: Different AI tools such as ChatGPT, Gemini, and Claude are becoming very popular. Although they are helpful for many day-to-day tasks, they can be use
arXiv:2605.17106v1 Announce Type: new Abstract: Production LLM deployments increasingly maintain heterogeneous model pools spanning order-of-magnitude cost differences. Existing routers make binary st
arXiv:2605.18583v1 Announce Type: cross Abstract: Coding agents now run autonomously with shell, file, and network privileges. When a user issues a benign request, the agent sometimes does more than a
rsi is here. jesus https://x.com/nickevanjoseph/status/2056760504949842219?s=46 Excited to welcome Andrej to the Pretraining team! He'll be building a team focused on using Claude to accelerate pretra
arXiv:2605.17062v1 Announce Type: cross Abstract: Spracklen et al. (USENIX Security '25) showed that code-generating large language models hallucinate package names that do not exist on PyPI or npm at
arXiv:2605.16699v1 Announce Type: new Abstract: Capped-usage SaaS products -- LLM subscriptions such as Claude Code and ChatGPT, cloud platforms such as Vercel and Cloudflare Workers, corporate benefi
Announcing coding harness tracing for observing, evaluating, and improving coding agent workflows across Claude Code, Cursor, Codex, GitHub Copilot, and Gemini CLI. The post Coding agent tracing and e
NEW paper worth reading. GPT-5.4 nano plus a critic-comparator orchestration loop hits 76.4% on SWE-bench Verified, matching standalone Gemini 3 Pro and Claude Opus 4.5 Thinking. The trick is to selec
Well, that's a Turing Test of a sort. (But gosh is the AI writing obvious if you use these systems at all - and this is obviously ChatGPT writing, not Claude) Well, this is a first: a ChatGPT-generate
a new era of hackathons: 2005→ look what i built with web search 2016 → can you build a rails app in 24 hrs? 2025 → spin up a CX bot with Claude in 5 min 2026 → train your own model over a weekend hap
Andon Labs conducted an experiment in which four major AI models (ChatGPT, Gemini, Claude, and Grok) were tasked with operating 24/7 radio stations continuously for five months. Each AI was required t
I like what Langchain recently released in their deepagents harness, which is an adapter to modify the syntax of primitive file system commands depending on the model Claude likes “Bash”, Gemini likes