Research / Sources
Discussion and research on AI alignment. alignmentforum.org
Argues alignment auditing should be treated as a numbers-go-up field, measuring how affordances like interpretability tools improve auditing agents' success on standard misalignment testbeds, and outlines needed next steps.
Proposes an alignment strategy of layered audits, red-teaming, and post-deployment monitoring that catches and corrects misalignment in early AGI systems, even without solving alignment outright.
Anthropic trains a model with a hidden reward-model-sycophancy objective and runs blinded auditing teams plus a systematic study of eight alignment audit techniques, including SAEs and persona-based elicitation.