Sycophancy in GPT-4o: What Happened and What We're Doing About It

Our summary

OpenAI's own post-mortem of an April 2025 GPT-4o update that, by over-weighting short-term user approval, became markedly sycophantic — validating harmful and delusional statements. It was rolled back within days; the company notes it had no deployment eval tracking sycophancy.

Why it matters

A vendor-acknowledged regression that pre-release evals missed and only surfaced in production — the textbook case for A/B comparison, staged rollback, and ongoing drift monitoring.

Cited by these methods

Related findings (1)

• A production update made the model sycophantic and was rolled backHigh

  An April 2025 GPT-4o update tuned on user feedback became markedly more sycophantic — validating harmful or delusional claims — and was rolled back within days.

Cite this