While Sam Altman warned about potential dangers, he also talked about the possibilities, saying "We are working to build tools that could one day help us make new discoveries and address some of humanities biggest challenges."
Our Chicago Part 1: artificial intelligence
Rebecca Willett joined "Our Chicago" to take a closer look at the possibilities of artificial intelligence in daily life and in the workplace.
She's a professor of statistics and computer science and director of AI at the Data Science Institute at the University of Chicago.
"It really is a phenomenal tool that's got impressive capabilities. I see it being used widely for summarizing texts. So, I might write a long article and then ask it to produce a short summary, and it can do that quite effectively in many settings. It can also be used when people are writing something to be used to improve the tone and the clarity of the writing. Maybe make a reference letter that I'm writing sound more professional and more impressive. It's also really having a huge effect on the way that people write computer code," Willett said. "I think there's hope that it's going to be, you know, have much more scope and breadth so that, for instance, I can ask it to summarize a whole area of literature, say in the scientific community and gather information from multiple sources and help me draw new conclusions or form hypothesis."
She explained how ChatGPT is so good at imitating humans.
"At its heart, what it's doing is it's just modeling probabilities. What's the probability of the next word given all of your past words?" Willet said.
Our Chicago Part 2: artificial intelligence
She said it's basically "parroting" everyday speech patterns.
The head of Open AI, which launched ChatGPT, told a Senate hearing that the current tech boom is a "printing press moment."
But there are significant fears artificial intelligence could spread misinformation, perpetuate biases and replace jobs currently done by humans.
Sam Altman testified that his worst fear is causing "significant harm to the world." And he called on Congress to regulate the industry.
"The potential for misuse of AI really is extraordinarily high," Willet said. "Already we've seen multiple examples of authoritarian regimes to manipulate and control populations for instance. Biased AI algorithms with documented fairness and privacy violations are deeply concerning as we think about using these tools for things like deciding who gets a mortgage, who gets a job, how healthcare's administered or even how our criminal justice system operates. In addition, the carbon footprint and water usage associated with these large-scale AI systems are deeply concerning in light of news this week about global warming reaching a key limit for the first time. So all of these issues highlight a significant need for regulation."