OpenAI Seeks To Remove Commercial ‘AGI’ Constraint

ChatGPT developer OpenAI is in discussions to remove a provision that bars commercial entities such as Microsoft from using its most powerful AI models once the start-up achieves “artificial general intelligence” (AGI), which the company believes could unlock billions in future investments, the Financial Times reported.

The discussions come as OpenAI prepares to restructure from a not-for-profit research lab to a public benefit corporation, a major step toward a profit-oriented structure that started when it created a for-profit subsidiary and began accepting outside investment from Microsoft in 2019.

Under current terms, when OpenAI creates AGI, defined in OpenAI’s current charter as a “highly autonomous system that outperforms humans at most economically valuable work”, Microsoft’s access to such a technology would be void.

The OpenAI board would determine when AGI is achieved.

Commercial incentive

The start-up is considering removing the stipulation from its corporate structure, allowing Microsoft to continue investing in the company and accessing all of its technology after AGI is achieved, the FT‘s report said, citing multiple sources.

It said a final decision has not been made and options are being discussed by the board.

The provision was included as part of OpenAI’s mission to benefit humanity and to prevent powerful technologies such as AGI from being misused for commercial gain, instead giving ownership of the technology to its non-profit board.

“AGI is explicitly carved out of all commercial and IP licensing agreements,” OpenAI says on its website.

But such a limit could disincentivise Microsoft from future investments in OpenAI, whose development of AI technologies consumes huge amounts of capital due to the intensive use of computing and electrical power that the work entails.

Microsoft has already invested $13 billion (£10bn) in OpenAI and a further round of funding from other firms recently valued the start-up at $150bn, but the company faces competition from deep-pocketed rivals such as Google, Amazon and Meta.

New structure

As part of OpenAI’s shift to a public benefit corporation, the company is discussing new terms with investors, including its largest shareholder Microsoft, the FT‘s report said.

OpenAI chief executive Sam Altman told a New York Times conference last week that if the firm had known how much capital it would need it “would have picked a different structure”.

The start-up has attracted criticism for its steady shift toward a for-profit focus, including removing constraints such as a prior focus on “safety”, a term used to describe the oversight of superintelligent AIs.

Many of the company’s founding team have left the company, including former chief technology officer Mira Murati, who left in September as the company was moving to ditch its non-profit structure.