OpenAI tightens security measures

OpenAI, a leading company in the field of artificial intelligence, has significantly increased its security operations against the risk of corporate espionage. The move comes after a China-based startup called DeepSeek launched a similar model in January.

OpenAI alleges that DeepSeek copied its models without authorization.

As part of the security measures, access for OpenAI personnel to sensitive algorithms and new products has been strictly limited.

Only verified team members involved in the project can access the o1 model developed by the company, and these discussions can only take place in designated office areas. Furthermore, proprietary technologies are now isolated on offline computers, and biometric access controls are being implemented in offices.

While OpenAI is increasing physical security at its data centers, it is also expanding its cybersecurity team.

These changes reflect concerns regarding threats from foreign competitors to OpenAI's intellectual property. However, internal security issues, such as the ongoing competition among American artificial intelligence companies and the leaking of comments by CEO Sam Altman, may also have been factors in the implementation of these measures.