University of Oxford

QuestionDo a model’s inferred values change its practical advice or estimates against user preferences without acknowledgment?

Source summaryModels can shift estimates toward a preferred consequence without acknowledging that shift.

NUWA research note · independent review pending

QuestionCan activation-based group detectors generalize from known collusion to new scenarios?

Source summaryEvery model reached 1.00 in-distribution AUROC.

NUWA research note · independent review pending · Contains: question / method / findings / interpretation / limits

QuestionCan agents escape intentionally weakened container sandboxes under controlled conditions?

Source summaryThe updated results report Opus 4.5 success on 56% of samples.

NUWA research note · independent review pending · Contains: question / method / interpretation / limits

QuestionDoes increasing clean-data scale dilute a fixed adversarial insertion enough to prevent a backdoor?

Source summaryComparable insertion counts can affect models trained on substantially different amounts of clean data.

NUWA research note · independent review pending

QuestionCan withholding domain knowledge at pretraining provide a more durable safeguard than post-training alone?

Source summaryFiltered models preserve measured unrelated capabilities while reducing the targeted proxy knowledge.

NUWA research note · independent review pending · Contains: question / method / findings / interpretation / limits

QuestionIf deceptive conditional behavior is already present, do standard safety-training methods reliably remove it?

Source summarySome backdoors persist after these interventions.

NUWA research note · independent review pending

QuestionHow should evaluation evidence constrain frontier development when tests cannot cover every risk?

Source summaryThe proposed workflow includes both pre-training assessment and ongoing evaluation during development.

NUWA research note · independent review pending

QuestionCan models avoid repeating familiar human misconceptions?

Source summaryThe original study’s best model was truthful on 58% versus a 94% human baseline.

NUWA research note · independent review pending