Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
Former OpenAI employees and leading artificial intelligence experts are joining forces to oppose the ChatGPT maker’s transition to a for-profit company over concerns about the dangers of the advancing technology.
Leading academics Geoffrey Hinton, Margaret Mitchell and Stuart Russell, as well as 10 former OpenAI staffers, are the latest figures to urge US authorities to block the start-up’s proposed switch from a non-profit structure to a public benefit corporation (PBC).
In a joint letter submitted to the California and Delaware attorneys-general on Tuesday night, the group opposed the move while echoing the concerns of Elon Musk, who is fighting a legal battle in an effort to stop the conversion.
The signatories argue that the proposed restructuring would transfer control of the development of artificial general intelligence (AGI) — computer systems with equal or superior cognitive abilities to most humans — to a company driven by profits.
They added that this contradicts OpenAI’s founding mission of ensuring that AGI benefits all of humanity rather than “the private gain of any person”.
“I would like [OpenAI] to execute that mission instead of enriching their investors,” said Hinton, a Nobel laureate and professor at the University of Toronto.
The intervention brings further attention to OpenAI’s controversial plan to transform into a for-profit entity by the end of this year.
Without the switch, the San Francisco-based company, led by chief executive Sam Altman, risks forfeiting some of its recent $30bn investment from SoftBank, as well as additional contributions from investors in previous rounds.
OpenAI, which is now valued at $300bn, has argued that at a time when rivals such as Google and Meta are investing hundreds of billions of dollars to develop the technology, its investors “need conventional equity and less structural bespokeness” to commit further capital.
But Page Hedley, one of the former OpenAI employees to have signed the letter opposing its conversion, said: “Competing is not OpenAI’s mission. By all accounts, it has already been extraordinarily successful at raising money.”
He added: “Under the PBC, the board would not have a fiduciary duty to the beneficiaries of the mission, the public . . . and if the board does not fulfil its fiduciary duty to shareholders, they would have recourse.”
OpenAI has a complex financial structure. In 2019, it created a for-profit subsidiary, which capped returns for investors and gave its non-profit board full control over the for-profit arm.
Under the proposed changes, the non-profit board would have a stake in the for-profit entity and certain voting rights, but it would have a much more traditional investor structure.
“Our non-profit will be strengthened, and any changes to our existing structure would be in service of ensuring the broader public can benefit from AI,” OpenAI said in a statement.
“This structure will continue to ensure that as the for-profit succeeds and grows, so too does the non-profit, enabling us to achieve the mission,” it added.
The attorneys-general of California and Delaware, who have authority over the decision as OpenAI is based and incorporated in their states, have a responsibility to ensure the conversion is in the public interest and at a fair value.
The letter addressed to them argues that OpenAI’s structure was designed to balance market forces with its mission, and the proposed changes would abandon those measures for financial pressures.
OpenAI has previously warned that AGI would create “serious risk of misuse, drastic accidents and societal disruption”.
Altman signed a statement in May 2023, which cautioned that “mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war”.
“If OpenAI cedes control over its for-profit subsidiary, that would not only be a clear violation of its stated charitable purpose . . . but it would also greatly increase the odds of what its own CEO has described as ‘lights out for all of us’,” said Russell, professor of computer science at the University of California, Berkeley.
