OpenAI has announced the creation of a ‘Preparation Team’ dedicated to assessing and forecasting potential risks associated with artificial intelligence (AI). This initiative is awaiting final approval from the company’s board of directors.
The newly formed team at OpenAI will serve as a bridge connecting the security and policy teams within the organization. It is tasked with devising measures and safeguards against the “destructive risks” posed by potentially uncontrollable powerful AI models. OpenAI also stated that certain technologies would only be utilized if deemed safe.
The proposed plans include the new advisory team reviewing security reports, which will then be forwarded to company executives and the OpenAI board for approval. While executives have the authority to make final decisions technically, the new plan grants the board the power to reverse security decisions.
The announcement follows a turbulent period for OpenAI, marked by the sudden dismissal of Sam Altman in November, his subsequent reinstatement as CEO, and the introduction of a new board including Bret Taylor, Larry Summers, and Adam D’Angelo.
OpenAI’s ChatGPT was released to the public in November 2022, sparking increased interest in AI. Concerns about potential societal dangers, which may have influenced Altman’s dismissal, are also circulating. In July, leading AI companies, including OpenAI, Microsoft, Google, and Anthropic, launched The Frontier Forum to facilitate self-regulation for responsible AI creation. Additionally, the Biden Administration released an executive order in October outlining new AI safety standards for high-level model development, and leading AI developers were invited to the White House to commit to developing safe and transparent AI models.