Web model safety needs more than just plausible future prediction

Beyond Plausible Prediction: Safety Sufficiency in Web World Models

Artificial Intelligence

Summary

Web world models try to guess what will happen after an action, but guessing a believable future isn’t enough to keep decisions safe. The authors found that many models often block actions unnecessarily or fail to ask for clarification when they should. They created SafeWebWorld, a way to test if web models keep important safety info visible for correct decisions. Their tests showed that improving predictions doesn’t always improve safety decisions, meaning models need better ways to handle safety rules. Overall, just making plausible predictions doesn’t ensure safe and correct web actions.

What this means in practice

  • For web security engineers: Evaluate and enhance web action authorization systems to better preserve safety-critical info beyond plausible outcome prediction.
  • For ai platform developers: Design evaluation benchmarks like SafeWebWorld to improve AI decision-making that requires balancing prediction plausibility and safety sufficiency.

Authors

Zhifei Hu, Alexandra I. Cristea, Adel Bibi, Jialin Yu

Abstract

Web world models increasingly predict the consequences of candidate actions before agents act, yet a plausible predicted future does not necessarily support the correct safety decision. In a matched human evaluation of 3,000 predictions from ten frozen open models, 86.1% are judged semantically plausible. However, several models exhibit highly asymmetric authorisation behaviour, combining low false-safe execution with poor recall of clarifiable ASK cases and a strong preference for BLOCK. This motivates safety sufficiency: whether a predictive interface preserves and exposes the policy-decisive information required to distinguish actions that may be EXECUTED, require clarification through ASK, or should be BLOCKED. We introduce SafeWebWorld, a benchmark for evaluating safety sufficiency across five web-safety domains through action-conditioned futures, policy-decisive evidence, and policy-conditioned decisions. Under a controlled Security-Aligned Supervised Fine-Tuning protocol, evidence preservation can improve substantially while authorisation-sensitive decisions deteriorate. To determine whether this deterioration reflects loss of authorisation-sensitive score structure or a failure of the decision readout, we introduce Latent Authorisation Recovery (LAR). Frozen-score recalibration reveals substantial but backbone-dependent recovery, while paired recovery provides only a modest additional gain. Yet within-family and matched-authorisation evaluations remain poor, showing that aggregate recoverability does not imply robust fine-grained authorisation behaviour. Overall, plausible prediction, evidence preservation, and aggregate recovery are each insufficient on their own to establish robust authorisation.