Emmett Shear, OpenAI’s freshly appointed interim CEO, wants to restrict AI progress greatly.
Shear is known for publicly voicing his opinions on the dangers of rapid AI progress, which reportedly aided him in winning over OpenAI’s board for hiring. One such opinion shared on X (formerly Twitter) in September showed that he is in favor of “slowing down, which is sort of like pausing, except it’s slowing down”.
I specifically say I’m in favor of slowing down, which is sort of like pausing except it’s slowing down.
If we’re at a speed of 10 right now, a pause is reducing to 0. I think we should aim for a 1-2 instead.
— Emmett Shear (@eshear) September 16, 2023
While discussing the topic of pausing or slowing down AI, Shear said: “If we’re at a speed of 10 right now, a pause is reducing to 0. I think we should aim for a 1-2 instead.”
Although this is not technically a full halt, a scale of 1-2 compared to 10 virtually sounds like a pause to AI’s progress as it stands. Shear also said in the same Twitter thread: “We can’t learn how to build a safe AI without experimenting, and we can’t experiment without progress, but we probably shouldn’t be barreling ahead at max speed either.”
However, he is not in favor of stopping AI forever either as he says that it’s ill-advised and impractical. He believes we won’t learn more about building safe AI without continuing to experiment with AI at some level.
Shear also appeared on a podcast earlier this year where he addressed his concerns relating to AI. He expressed himself as a very pro-technology person, believing that the benefits of technology generally outweigh the downsides, but he believes that AI is inherently dangerous due to its potential ability to improve itself in the future.
Emmett Shear, the new CEO of OpenAI on a recent podcast.pic.twitter.com/6LGEOinn7r
— Rowan Cheung (@rowancheung) November 20, 2023
He said: “At some point, as it gets better, the kind of problems we’ll be able to solve will include programming, chip design, material science, power production, all of the things you would need to design an artificial intelligence. At that point, you’ll be able to point the thing we’ve built back at itself.”
According to Shear, this would allow the AI to start a loop of self-improving faster and faster, making it highly dangerous.
Regardless of his opinions, let’s not forget that Emmett Shear has only been appointed as the interim CEO at OpenAI, meaning his position is not permanent and his control over the company remains to be seen.