OpenAI publishes LifeVal, a benchmark of everyday tasks
On 2 November 2026 OpenAI published LifeVal. The page shows workflow examples, not a benchmark average. The new survey had 3,995 respondents.
Anthropic publishes FlashProve, a prototype for provable inference
Anthropic's page, dated 2 November 2026, introduces FlashProve. In one case, an H100 generates proofs at about 110 tokens per second. It is a research prototype, not a product.
The Frontier Model Forum publishes a framework for rating cyber jailbreaks
On 6 November 2026 OpenAI announced a Frontier Model Forum framework for comparing cyber jailbreaks. This note does not describe a technique. Public comments run through 21 December 2026.
Anthropic describes Claude's memory, and the step it calls dreaming
On 6 November 2026 Anthropic published an engineering post on agent memory. Dreaming is the offline pass that reorganizes logs between sessions. It is not a weight update.
BioBasicsBench measures agents on routine wet-lab tasks
On 6 November 2026 OpenAI published BioBasicsBench with Ligo. One hundred tasks test whether an agent can read raw lab evidence, spot an error, and stop when the evidence does not support a conclusion.
DeepMind says agents found two weaknesses in legacy 1024-bit RSA
On 6 November 2026 Pushmeet Kohli said human-guided agents found two weaknesses in legacy 1024-bit RSA. He wrote that RSA 2048 is not affected. This desk did not open the papers.
Karpathy ships TristanTube, a filter of 47,000 videos
On 7 November 2026 Andrej Karpathy posted TristanTube, a filter built from Tristan Hume's YouTube tier list. The 47,000 count and the three-minute Claude Code claim are in the post, not in the page HTML this desk extracted.
GPT-6 Cyber stays in Daybreak Red
OpenAI announced GPT-6 Cyber on 5 November 2026. Select Daybreak Red customers get it through the API and Codex. There is no public release.