\n<\/p>\n
OpenAI has reportedly overhauled its security operations to protect against corporate espionage. According to the Financial Times<\/a>, the company accelerated an existing security clampdown after Chinese startup DeepSeek released a competing model in January, with OpenAI alleging that DeepSeek improperly copied its models using \u201cdistillation\u201d techniques.<\/p>\n The beefed-up security includes \u201cinformation tenting\u201d policies that limit staff access to sensitive algorithms and new products. For example, during development of OpenAI\u2019s o1 model, only verified team members who had been read into the project could discuss it in shared office spaces, according to the FT.<\/p>\n And there\u2019s more. OpenAI now isolates proprietary technology in offline computer systems, implements biometric access controls for office areas (it scans employees\u2019 fingerprints), and maintains a \u201cdeny-by-default\u201d internet policy requiring explicit approval for external connections, per the report, which further adds that the company has increased physical security at data centers and expanded its cybersecurity personnel.<\/p>\n The changes are said to reflect broader concerns about foreign adversaries attempting to steal OpenAI\u2019s intellectual property, though given the ongoing poaching wars<\/a> amid American AI companies and increasingly frequent leaks<\/a> of CEO Sam Altman\u2019s comments, OpenAI may be attempting to address internal security issues, too. <\/p>\n We\u2019ve reached out to OpenAI for comment.<\/p>\n<\/div>\n