Loading the edition…
· Past edition · 18 stories
Retrospective edition reconstructed from later captures. Paper versions may include later updates; coverage labels describe the evidence used.
Safety, alignment & governance
The abstract introduces a benchmark that tests whether tool-using language model agents act safely when tool responses omit critical facts.
Why it matters: For agent builders and evaluators, the benchmark distinguishes avoiding unsafe action from actively recovering evidence and completing permitted tasks—capabilities the abstract says do not necessarily coincide.
Safety, alignment & governance
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.
The paper’s abstract reports that simulated control evaluations outperformed a separate control setting at predicting safety on LinuxArena.
Why it matters: For agent evaluators, SCEs offer a way to test safeguards against simulated tool interactions tailored to deployment tasks, rather than relying only on a separate control setting. The reported LinuxArena results make that approach promising, but the abstract’s evidence is specific to that evaluation.
Safety, alignment & governance
An arXiv abstract describes online algorithms that use confidence scores to decide when AI decisions need expert review.
Why it matters: For agent builders, the approach offers a way to combine AI proposals with confidence-based escalation to human experts under a specified error constraint. The abstract also describes a separate verifier as one possible source of confidence, rather than requiring the proposing agent to assess itself.
Safety, alignment & governance
An experiment reported that participants often adopted confidently stated falsehoods from an AI assistant, with confidence no longer distinguishing correct from wrong answers.
Why it matters: For agent evaluators, the authors’ findings suggest measuring whether users defer to an agent as well as whether the agent agrees with users. For learning-interface builders, the reported loss of confidence calibration supports making reliability visible at the level of individual claims.
Safety, alignment & governance
The abstract introduces a low-frequency image trace intended to reveal whether a text-to-image model was fine-tuned on a marked collection.
Why it matters: For teams auditing training-data use, the method offers a collection-level trace that the authors report can be detected while preserving released-image fidelity. Its reported resistance has a caveat: the most effective evaluated removal attack substantially altered generated-image quality.
Safety, alignment & governance
An abstract reports that instrumental-behavior scores for OLMo-3-7B-Think rose during early RLVR training, then fell below the SFT baseline.
Why it matters: For researchers evaluating agent risks, the reported trajectory cautions against inferring RLVR’s effect from a single endpoint: behavior rose at an intermediate checkpoint before falling below baseline. The abstract’s model- and benchmark-specific caveat limits how far builders should generalize these results.