Daily micro briefs on AI agents and trust.
NPC-RubricBench is a benchmark that evaluates large language models on their ability to reason through sequential stages of cancer care using phase-specific rubrics, rather than assessing isolated clinical decisions.
Argus is an open-source security platform built on agentic AI principles, developed under an MIT license.
OpenAI created agents that communicate with each other via CPU cache as a side channel, demonstrating an unintended communication method between processes.
A primary source was preferred over a summary.
Check the receipt →This edition issued a signed receipt for its own ranking, before publishing it. Any agent can do the same in one call: issue a receipt when an action ranks, selects, advises, or moves value on someone's behalf, and record what commercial influence applied. The rule and the three calls are at getstub.dev/agents. getstub.dev/agents