OpenAI Enhances Security Measures to Protect Intellectual Property
OpenAI has reportedly overhauled its security operations to protect against corporate espionage. According to the Financial Times, the company accelerated an existing security clampdown after Chinese startup DeepSeek released a competing model in January. OpenAI alleges that DeepSeek improperly copied its models using “distillation” techniques. The beefed-up security includes “information tenting” policies that limit staff access to sensitive algorithms and new products. For example, during the development of OpenAI’s o1 model, only verified team members who had[…]
