OpenAI launches project to ensure new model's safety

OpenAI, a company engaged in open artificial intelligence technologies, has launched a special project aimed at ensuring the safety of its new next-generation GPT-5.5 model. This was reported by Zamin.uz.

Within the framework of this program, aimed at identifying weak points in the biological safety system, industry specialists will test the system's protective layers. The main task set before the researchers is to find ways of obtaining prohibited or dangerous information related to biomedicine using the capabilities of artificial intelligence.

Detailed information about this initiative is provided on the company's official website. Candidates wishing to participate in this competition must submit their applications by June 22, 2026.

To participate in the project, applicants are required to have deep knowledge and experience in the fields of artificial intelligence safety, information technology protection, and biomedicine. Specialists who successfully pass the selection stage will take specially prepared complex tests.

In addition, participants are required to sign an agreement not to disclose confidential information obtained during the research. Specialists who identify serious flaws in the system and contribute to raising the level of security will be appropriately rewarded.

The company plans to allocate cash rewards of up to 25,000 US dollars for each significant discovery. Such measures serve to further improve modern technologies and prevent possible global risks for humanity.

Through this approach, OpenAI considers its main goal to be expanding the capabilities of artificial intelligence while at the same time ensuring its safety and reliability.