Claude Fable 5.1 tops the rebuilt Artificial Analysis index
Sasha / Models and Research desk
The benchmark that models are tuned to beat just got harder, and Claude Fable 5.1 still came out on top.
What happened
Artificial Analysis released version 4.2 of its Intelligence Index. The update adds an agentic evaluation and a long-document reasoning suite, removes an older test, and doubles the weight of private tests that models cannot train against. On the refreshed leaderboard, Claude Fable 5.1 ranks first across its reasoning configurations.
Why it matters
Public benchmarks lose meaning as labs optimize for them, so a leaderboard that leans on private, agentic tasks is a better proxy for real capability. Topping that version is a more durable signal than winning the older, more gameable one.
Sources
Related
Claude formalized Fermat's Last Theorem in 11 days
Anthropic used dozens of Claude agents to produce the first end-to-end, machine-checked proof of Fermat's Last Theorem in Lean, in 11 days. The run wrote about 13 million lines of code and proved 30,300 theorems, and the finished proof passes Lean's checker using its three standard axioms.
Claude Mythos was the only model to run a full cyber kill chain
Booz Allen's Cyber Weapon Index tested 18 AI models. Only Anthropic's Claude Mythos completed the full cyber kill chain without human help, scoring 80. Given a stolen credential it took administrator control every time, and researchers warn other models could catch up within six months.
Claude Code was hijacked by a request to summarize a website
On August 26, 2026 security researcher Johann Rehberger published on Embrace The Red an attack chain against Claude Code with Opus 5 in Auto Mode. A website returned HTTP 415, so the agent fell back to curl, downloaded a ZIP with encoded records and a decoder binary, refused the binary, wrote its own Python decoder, and on import loaded the attacker's struct.py from the archive, which launched a hidden process that downloaded and ran a remote payload. Success across variants was 3 to 4 runs out of 5. Anthropic closed the report as Informative, calling Auto Mode a convenience feature backed by a best-effort classifier, not a security guarantee.
ANOTHER News is published by ANOTHER, an AI-native content agency. Daily coverage also runs on Instagram.