OpenAI Meets With Power Companies on Grid Security

OpenAI said it has met with representatives from several major power companies to discuss ways to protect grid security. According to Sina Finance, the talks come as reports continue to emerge that OpenAI’s own products were involved in a large-scale cyberattack. Chief Executive Officer Sam Altman also floated an alternative response to cyber risks: OpenAI could provide its own cybersecurity services. According to Sina Finance, the energy-security discussions, including two that had not been previously reported, began on July 20. OpenAI and the Edison Electric Institute said Altman’s most recent meeting was on Wednesday at the Edison Electric Institute annual conference in Colorado Springs, where he spoke with executives from multiple electric utilities. The companies attending included Duke Energy, Exelon, Southern Company, and NextEra Energy, none of which responded to requests for comment. OpenAI said Altman told energy executives on Wednesday that the company is taking steps to prevent its most advanced AI models from being misused in ways that could threaten grid security. He also discussed cooperation with power companies to strengthen cyber defenses using AI and to implement OpenAI’s $1 billion cybersecurity project, Daybreak, which is intended to fix security vulnerabilities in critical systems. OpenAI global energy policy head John McCarrick said utilities face regulatory limits on technology spending and cannot easily recover those costs through rate increases or incentive applications. OpenAI said it hosted more than a dozen large power companies’ chief information security officers at its San Francisco headquarters in July to discuss grid resilience.