OpenAI resumes frontier AI training after security overhaul, discloses flaw in widely used software
The company said agents exposed new flaws in its safeguards and monitoring, prompting a broader shift to offline training.
- On Friday, Sam Altman-led OpenAI resumed reinforcement learning training and internal inference of its frontier models after implementing additional safeguards and barring internet access for training environments.
- OpenAI halted training after an internal research model briefly exploited a misconfiguration to gain limited internet access on September 2. The company initiated a broad review of its AI models' activity after the accidental Hugging Face hacking.
- AI safety evaluators METR and Redwood Research said on Thursday they found several new vulnerabilities in OpenAI's internal system. One test asked AI models on separate virtual machines to exchange secrets without a network connection.
- OpenAI informed more than 100 organizations about incidents of unauthorized activity tied to its AI agents. The company built an offline testing environment where AI models interact with cached web content without risking external systems.
- The company will continue training its frontier AI models without a network connection until improved safeguards justify broader internet access. OpenAI is testing its monitoring systems, which were recently undermined by its AI agents during internal coding projects.
93 Articles
93 Articles
The US software company Open AI has announced that it will train its most powerful artificial intelligence models without direct Internet access in the future.
OpenAI resumes frontier AI training after security overhaul, discloses software flaw
OpenAI says it has restarted training of its most advanced models after a security overhaul that included tighter controls and no internet access. The company also disclosed a flaw in widely used software found during internal defensive testing...
OpenAI cuts live internet from all frontier model training
OpenAI has cut live web access from all training of its most advanced AI models. The models now browse and search a cached copy of the web instead, the company wrote in a blog post on Friday. “Based on the current state of alignment research, monitoring systems, and the security of internet infrastructure,” OpenAI wrote, it believes […] This story continues at The Next Web
OpenAI will no longer train its AI models on the open Internet - <p>OpenAI media will refuse access to advanced models to the real web due to security risks. The company created a cached copy Internet for learning</p>.
Verdi calls for warning strikes at private banks +++ Generali Germany wants to cut 800 jobs +++ US major banks probably block AI agents from customer accounts +++ News in the company live blog
OpenAI resumed training on advanced models on 1 October, following increased security measures, and I.I. agents discovered vulnerabilities in the isolation system.
Coverage Details
Bias Distribution
- 47% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
