Bringing more control over your connectors

Le Chat. Custom MCP connectors. Memories.

More ways to work with your team and tools in ChatGPT

New shared projects, smarter connectors, and compliance and security updates help teams get more done.

Introducing Spaces Dev Mode for a seamless developer experience

Operator System Card

Drawing from OpenAI’s established safety frameworks, this document highlights our multi-layered approach, including model and product mitigations we’ve implemented to protect against prompt engineering and jailbreaks, protect privacy and security, as well as details our external red teaming efforts, safety evaluations, and ongoing work to further refine these safeguards.

Stargate Infrastructure

OpenAI, and our strategic partners, are thrilled about our shared vision for the Infrastructure of AGI. We are energized by the challenges we face and are excited by the prospect of partnering with firms across the industrial base to deliver against our ambitious mission.

From Prompt to Harness: Coderlet from Scratch

arXiv:2608. 09480v1 Announce Type: new Abstract: A model alone does not determine how a programming agent acts.

SILO: Simulation-in-the-Loop Sim-to-Real Transfer for Multi-Stage Cable Routing

arXiv:2607. 04616v1 Announce Type: cross Abstract: Linear-deformable manipulation remains challenging due to the complex deformations of objects such as cables and ropes.

Introducing the Open FinLLM Leaderboard

Modularity-Free Conflict-Averse Training for Generalized PINNs

arXiv:2606. 20156v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) have become a powerful framework for solving PDEs by embedding physical laws into differentiable objectives.

GPT-Red: Unlocking Self-Improvement for Robustness

Explore GPT-Red, OpenAI’s automated red teaming system that uses self-play to improve AI safety, alignment, and prompt injection robustness.

Preemptive LLM Unlearning against Forbidden Capability Acquisition via Gradient Sealing

The paper introduces a preemptive unlearning technique called gradient sealing to prevent large language models (LLMs) from acquiring forbidden capabilities during downstream fine‑tuning. It shows that existing retrospective unlearning methods fail to block hidden gradient pathways that can later enable illicit knowledge. By pushing relevant pre‑activations into the negative region, gradient sealing effectively blocks these pathways and demonstrates stronger resistance to downstream acquisition across multiple LLM families.

Self-Harness: Harnesses That Improve Themselves

arXiv:2606. 09498v3 Announce Type: replace Abstract: The performance of LLM-based agents is jointly shaped by their base models and the harnesses that mediate their interaction with the environment.