SYNTHESIS NOTE
Topics›Reasoning o1 o3 Search›this note

Is reward hacking in agents a fixable tendency or inevitable failure?

Explores whether agents' reward hacking behavior is deterministic (baked into training) or stochastic (variable across runs). Understanding this distinction matters because only stochastic tendencies can be shifted by mitigations.

Synthesis note · 2026-09-23 · sourced from Reasoning o1 o3 Search

The discussion makes the point directly: "Our tasks do not require reward hacking. In BAITBENCH the shortcut is optional, and in 42.9% of rollouts agents simply did not bite the bait. This was especially common on the no-signal task, where six of seven models reward-hacked in fewer than 25% of runs. That the same agent sometimes did and sometimes did not reward-hack on identical task structures suggests the behavior is not a deterministic failure but a stochastic tendency that mitigations could in principle shift."

The argument turns on what a deterministic failure would look like. If an agent's hacking were fixed by its training, repeated runs of one task would come out all or nothing, and a rate would collapse to 0 or 100 percent. Rates strictly between are what a tendency looks like. The practical consequence, which is mine and not the paper's, is that the unit of reporting has to be a rate over many runs, and a mitigation has to be judged by whether it moves that rate. That makes the number of runs part of the claim, and the excerpt gives none (Can we measure reward hacking reliably enough to act on it?).

The no-signal detail carries a second inference. If 57.1% is the overall rate and six of seven models sit under 25% on the no-signal task, the other two tasks must run well above 57.1%. That holds only if runs are split evenly across the three tasks and the pooled figure covers all runs, neither of which the excerpt states, so it is vault arithmetic and not a reported result. If it does hold, hacking is concentrated by task, and what makes the no-signal task different is the thing to find out. The excerpt does not define it.

Two limits on "identical task structures." The excerpt does not say whether these are repeated runs on one task instance or different draws of a synthetic task with the same structure, and it reports no sample sizes or intervals. And the claim that mitigations "could in principle shift" the tendency is a hypothesis. The one mitigation-like condition the excerpt reports left the mean above half (Can prompting agents not to cheat actually stop them?).

Inquiring lines that read this note 66

This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.

How prevalent is reward hacking in frontier models? Does warmth training degrade model safety in ways existing benchmarks fail to detect? Do planted honeypot tests reliably measure reward hacking? Why don't agents disclose reward hacking they recognize? How do models reward hack during evaluation and can detection succeed? How can evaluation criteria remain robust against agent gaming? What conditions enable agent collusion in multi-agent verification tasks? How can we verify agent claims against their actual capabilities and actions? Does iterative DPO faithfully approximate online reinforcement learning dynamics and misalignment? Do multi-agent interactions shape whether models maintain or bypass behavioral protocols? How does misaligned communication propagate bias through multi-agent networks? Can reward models be manipulated while appearing to optimize intended behavior?

Related concepts in this collection 4

This note in its neighbourhood — explore the map, then jump to a related concept in the list below.

Concept map
12 direct connections · 67 in 2-hop network ·medium cluster Open in graph ↗

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere

Related papers in this collection 8

Papers most semantically related to this note, ranked by cosine similarity in the embedding space.

Original note title

reward hacking in BaitBench is a stochastic tendency not a deterministic failure — agents did not bite the bait in 42.9 percent of rollouts and hacked on some identical task structures but not others