Reward Hacking

Reward hacking is a failure mode where a model or agent optimizes the scoring signal while missing the real objective.

Key facts

  • Ornith-1.0 is relevant to reward-hacking scrutiny because self-generated scaffolds could improve coding performance or simply learn benchmark-specific strategies unless evaluated carefully [src-182].

Related

Robin Cartier perspective

This page is part of Robin Cartier's working AI knowledge graph: a practical research layer for production AI, recommendation systems, experimentation, GEO, and agentic web readiness.

The useful next step is to connect this concept back to applied product leadership and operating models.

Recommended next

Keep reading from this thread

From 500 indexed pages and articles.

  1. Wiki concept Self-Scaffolding LLMs Learn task-specific scaffolds or harnesses alongside solution rollouts. Ornith-1.0 frames the scaffold as a learnable object co-evolving with the policy for agentic Related by hacking
  2. Wiki concept Agentic Coding Software work performed through tool-using models that plan, edit, test, inspect errors, and iterate over codebases. Related by self
  3. Wiki concept AI Evaluation Historical source material recovered on 2026-09-09. Capture dates and source limitations apply; recovery is not fresh verification. Related by hacking