****

OpenAI is undergoing significant structural changes, marked by the disbandment of its "AGI Readiness" team. This team was responsible for advising on OpenAI's ability to manage advanced artificial intelligence that could match or surpass human intellect and assessing global readiness for such technologies. The announcement came from Miles Brundage, the senior advisor for AGI Readiness, who also revealed his departure from the company in a Substack post on Wednesday.

Brundage cited high opportunity costs and the potential for greater impact in an external capacity as reasons for his exit. He asserted his desire to avoid bias and mentioned that he has accomplished his objectives at OpenAI. Brundage expressed concerns about the readiness of both OpenAI and other leading AI laboratories in his final remarks. He envisions starting a nonprofit or joining an existing one focused on AI policy research and advocacy, emphasizing that only a dedicated effort can ensure AI's safety and benefits are maximized.

This latest development follows the earlier disbandment of OpenAI’s Superalignment team in May, just a year after its formation. The Superalignment team aimed to develop scientific and technical solutions to control superintelligent AI systems. Multiple executive departures have accompanied these structural shifts, including CTO Mira Murati, research chief Bob McGrew, and research VP Barret Zoph, all leaving on the same day in September.

In recent uncertainty, OpenAI closed a significant funding round, pushing its valuation to $157 billion. This includes $6.6 billion from various investors and a $4 billion revolving line of credit, boosting the company’s liquidity to more than $10 billion. Despite these financial strides, OpenAI anticipates around $5 billion in losses against $3.7 billion in revenue this year.

Safety has become a critical issue for OpenAI, leading to the formation of an independent board oversight committee to review the company’s processes and safeguards. The 90-day review period culminated in recommendations released publicly. These changes follow growing safety concerns and controversies as OpenAI competes in the generative AI sector—a rapidly expanding market projected to exceed $1 trillion in revenue within the next decade.

In July, notable internal safety executive Aleksander Madry was reassigned, highlighting ongoing shifts within OpenAI's focus areas. This change came amid inquiries from Democratic senators regarding OpenAI's safety practices. Further, Microsoft relinquished its observer board seat in July, signaling confidence in the revised board structure post-Altman's brief ouster and subsequent reinstatement.

Internal and external scrutiny of AI industry practices persists, evidenced by a June letter from current and former employees addressing the rapid advancement of AI technology without adequate oversight. They called attention to the insufficient whistleblower protections and the reluctance of AI companies to share critical safety information willingly.

While OpenAI has taken strides to address security and governance issues, the dissolution of pivotal teams and key departures underscore the ongoing difficulties in balancing innovative growth with responsible stewardship of AI technologies.