With a safety-focused agenda, OpenAI had launched a new, independent board committee as the pressure mounted on AI to be safer.
In March 2024, OpenAI revealed an independent board oversight committee dedicated solely to safety and security. The move came as pressure mounted on AI to be safer, with the latest being fears over the biological war potential of weapons eased by AI models. Under this new move, the governance of OpenAI would shift to greatly favor independence; even the company’s CEO Sam Altman was left out sitting on the Safety and Security Committee, led by Carnegie Mellon University’s own Zico Kolter.
In part, this restructuring followed a closer examination of the firm’s safety protocols and governance; this review has further expanded into the safety and security processes of OpenAI. The newly formed committee oversees major safety evaluations for AI models and has the ability to put a halt to model launches if there is a discovery of emerging safety concerns.
It comes at a time when the company has been in the line of criticism concerning its safety policies, where it has been accused of opposing robust AI regulation before considering it. Criticism for such reasons has been heightened by the the exit of some of its concerned employees over the long-term risks of AI and has served to shape the decision for OpenAI to fortify its governance frameworks toward becoming more relevant with the mission of developing beneficial AI.
The new safety committee is also another part of the efforts by OpenAI to increase the security measures, make activities more transparent, and cooperate with external organizations such as governmental and independent AI safety institutes in the US and UK, which can help further research on AI safety standards and trustworthy AI practices.
