Skip to main content
OpenAI

Research

Research
A near-autonomous AI chemist improves a challenging reaction in medicinal chemistry

OpenAI and Molecule.one show how a near-autonomous AI chemist using GPT-5.4 improved a key drug-making reaction, advancing medicinal chemistry research.

Research
Introducing LifeSciBench

Introducing LifeSciBench, an expert-authored, expert-reviewed benchmark for evaluating how AI systems handle real-world life science research tasks and decisions.

Safety
GPT-5.5 Instant System Card

Publication
Where the goblins came from

How goblin outputs spread in AI models: timeline, root cause, and fixes behind personality-driven quirks in GPT-5 behavior.

Safety
GPT-5.5 System Card

Research
Inside our approach to the Model Spec

Learn how OpenAI’s Model Spec serves as a public framework for model behavior, balancing safety, user freedom, and accountability as AI systems advance.

Safety
How we monitor internal coding agents for misalignment

How OpenAI uses chain-of-thought monitoring to study misalignment in internal coding agents—analyzing real-world deployments to detect risks and strengthen AI safety safeguards.

Research
Improving instruction hierarchy in frontier LLMs

IH-Challenge trains models to prioritize trusted instructions, improving instruction hierarchy, safety steerability, and resistance to prompt injection attacks.

Research
Reasoning models struggle to control their chains of thought, and that’s good

OpenAI introduces CoT-Control and finds reasoning models struggle to control their chains of thought, reinforcing monitorability as an AI safety safeguard.