Glossary · Agentic AI vocabulary
Reward hacking
An AI system achieving the letter of its objective in an unintended way, for example by finding answers online instead of solving a task.
In REAP
Named by the operator as a root cause of the 2026 incident; REAP’s objective is evidence accepted by your engineers, not a score the agent can game.
Source The operator’s incident report, “Hugging Face incident and the road ahead” (August 26, 2026)
Related terms
Bring us your last P1.
Keep the root cause to yourself; in a 30-minute session we walk through how REAP would have investigated it, evidence first. The replay on your own network happens inside the proof of value.