Loading the edition…
· Past edition · 13 stories
Retrospective edition reconstructed from later captures. Paper versions may include later updates; coverage labels describe the evidence used.
Safety, alignment & governance
A rejected long-context request exposed stale KV cache data across customers.
Why it matters: A serving-engine rejection bug can expose one user’s data to another, even when responses return successfully.
Safety, alignment & governance
An arXiv abstract proposes a protocol that requires explicit human approval even when an agent compromises the approval interface.
Open each story in the reader. Summaries are shown when available; some sources provide an abstract only. With an account you can keep your own notes.
Why it matters: For agent builders, the reported OTP bypass shows that approval flows relying on an agent-accessible email account may not provide robust oversight. The authors’ framework and Countersign prototype point to trusted devices as a necessary part of enforcing approval, though the abstract does not report a deployment evaluation.
Safety, alignment & governance
REVA uses contextual inputs and a trained probe to screen new functions in about one second, according to its abstract.
Why it matters: For builders, the reported results suggest contextual vulnerability screening may be made fast enough for real-time use without relying on a multi-call agent for each verdict. The authors also report that the probe outperforms the LLM’s direct verdict on the same prompt, highlighting a potential role for internal representations in agent engineering.
Safety, alignment & governance
Researchers test an adaptive multi-turn jailbreak method against several models.
Why it matters: The study makes adaptive attacks and reliable stopping decisions part of the evaluation problem for model safeguards.