'Godfather of AI' Leaves Google to Voice Concern About Artificial Intelligence

pcmag.com:

Another well-known AI technologist is speaking up about potential pitfalls of AI.

"It is hard to see how you can prevent the bad actors from using it for bad things,” Geoffrey Hinton tells(Opens in a new window) The New York Times. "I console myself with the normal excuse: If I hadn’t done it, somebody else would have."

Hinton, 75, resigned from Google last month after more than a decade in order to "freely speak out about the risks of AI," the Times says. He won a 2018 Turing Award for pioneering some of the technology that now powers ChatGPT and similar generative AI tools—and has earned the nickname "Godfather of AI"—but now says a part of him regrets his life's work.

Hinton's immediate concern is the spread of misinformation: a flood of fake photos, videos, and text that makes it impossible to know "what is true anymore." New tools that create deepfakes, or computer-generated clones of a famous person's voice and appearance, make it easy for anyone with a computer and internet connection to misrepresent public figures.

Midjourney, for example, recently paused free access to its AI image generator amid a spike in demand. Some people were using the platform to generate fake images of public figures like Donald Trump (see below), Elon Musk, and the Pope.

Hinton also says AI will replace jobs at an alarming rate. He acknowledges most of these positions are "drudge work," but AI "might take away more than that," he says.

In the long term, he fears machines will begin to exhibit unplanned behavior, like writing and executing original code on their own. The most apocalyptic outcome of this could come in the form of autonomous weapons—killer robots—which Hinton strongly opposes. (US lawmakers recently moved to make sure AI