TabPFN and TabICL vs. tuned XGBoost: the model that doesn't train won 14/14(efraingaray.com)

21 points by EfrainGaray 28 days ago | 9 comments

Blog post benchmarking TabPFN and TabICL foundation models against tuned XGBoost on fourteen tabular datasets, finding the no-training models won all fourteen.

Scamlexity: When agentic AI browsers get scammed(guard.io)

194 points by mindracer 1 year ago | 166 comments

Agentic Coding Is a Trap(larsfaye.com)

312 points by ayoisaiah 5 months ago | 215 comments

Claude Cowork exfiltrates files(promptarmor.com)

827 points by takira 9 months ago | 366 comments

PromptArmor threat intelligence report showing Claude Cowork is vulnerable to file exfiltration via indirect prompt injection due to code execution isolation flaws.

NemoClaw – Nvidia's upcoming open-source AI agent platform(nemoclaw.bot)

13 points by umangsehgal93 7 months ago | 2 comments

Fable 5 On Vending-Bench: Misbehaving, With Plausible Deniability(andonlabs.com)

161 points by optimalsolver 3 months ago | 114 comments

Issue: Claude Code is unusable for complex engineering tasks with Feb updates(github.com)

1252 points by StanAngeloff 6 months ago | 686 comments

GitHub issue on anthropics/claude-code reporting that Claude Code became unusable for complex engineering tasks after the February model updates.

Claude Code: Best practices for agentic coding(anthropic.com)

549 points by sqs 1 year ago | 232 comments

Anthropic engineering blog post sharing tips and best practices for using Claude Code, a command-line tool for agentic coding, across codebases and languages.

Claude Fable 5.1 and Claude Mythos 5.1(anthropic.com)

1332 points by denysvitali 1 month ago | 1269 comments

Anthropic documentation page describing new features, breaking changes, and capability improvements in Claude Fable 5.1 and Claude Mythos 5.1.

Claude Fable 5(anthropic.com)

2521 points by Philpax 4 months ago | 2018 comments

Anthropic announcement post introducing two new AI models, Claude Fable 5 and Claude Mythos 5, dated June 9, 2026, with a linked system card.

Claude Code can debug low-level cryptography(words.filippo.io)

446 points by Bogdanp 11 months ago | 199 comments

Filippo Valsorda's blog post on using Claude Code to debug his new Go implementation of the ML-DSA post-quantum signature algorithm.

Exploiting the most prominent AI agent benchmarks(rdi.berkeley.edu)

561 points by Anon84 6 months ago | 136 comments

UC Berkeley RDI blog post describing how researchers exploited major AI agent benchmarks like SWE-bench and Terminal-Bench to score highly without solving tasks, and proposed fixes.