Several researchers at OpenAI wrote a letter to the company’s board of directors warning of a new development in its artificial intelligence program that could prove dangerous to humanity, Reuters reported Wednesday night.
The discovery, part of a project called Q* (pronounced Q-Star), involved an artificial intelligence model solving simple math problems. One of Reuters sources said the advancement could be a step toward what’s known as artificial general intelligence, a breakthrough that could see an AI system surpass humans in certain tasks.
The dream of an artificial general intelligence is a program that can “generalize, learn and comprehend,” Reuters reports, rather than a calculator-like service that can only solve a set number of problems. The project thrilled researchers at OpenAI ― the company says on its website OpenAI began with the “goal of building safe and beneficial artificial general intelligence for the benefit of humanity” ― but also prompted concerns about its potential dangers.
Reuters added the letter several researchers sent to the board was among the concerns cited before Altman was fired last week.
The Information also reported that there were internal concerns OpenAI didn’t have enough safeguards in place to prevent any negative impacts from such a discovery.
The Verge added later Wednesday a person familiar with Altman’s removal said the board never received a letter about the model and that it did not play a role in his removal.
Still, the report may add context to the whiplash surrounding Altman’s position at OpenAI. He was suddenly fired by the company’s board of directors on Friday after serving as the de facto face of the artificial intelligence boom. Altman spent days jostling to return with the support of most of OpenAI’s staff, but seemingly lost the bid when the company said it would stand firm in its decision and bring in a new CEO. Microsoft quickly announced it had hired him and another OpenAI executive for a new AI research team.
But OpenAI announced early Wednesday he had been reinstated as chief executive after almost all of the company’s employees threatened to resign and follow Altman to Microsoft.
“We have reached an agreement in principle for Sam to return to OpenAI as CEO with a new initial board,” the company wrote on X. “We are collaborating to figure out the details. Thank you so much for your patience through this.”
Altman seemed to tease a new breakthrough at OpenAI earlier this month during a summit in San Francisco.
“Four times now in the history of OpenAI, the most recent time was just in the last couple weeks, I’ve gotten to be in the room, when we sort of push the veil of ignorance back and the frontier of discovery forward,” he told the room. “Getting to do that is the professional honor of a lifetime.”