What if Dario Amodei Is Right About A.I.?

The Ezra Klein Show

New York Times Opinion

Society & Culture, Government, News

4.6 • 11K Ratings

🗓️ 12 April 2024

⏱️ 92 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

Back in 2018, Dario Amodei worked at OpenAI. And looking at one of its first A.I. models, he wondered: What would happen as you fed an artificial intelligence more and more data? He and his colleagues decided to study it, and they found that the A.I. didn’t just get better with more data; it got better exponentially. The curve of the A.I.’s capabilities rose slowly at first and then shot up like a hockey stick. Amodei is now the chief executive of his own A.I. company, Anthropic, which recently released Claude 3 — considered by many to be the strongest A.I. model available. And he still believes A.I. is on an exponential growth curve, following principles known as scaling laws. And he thinks we’re on the steep part of the climb right now. When I’ve talked to people who are building A.I., scenarios that feel like far-off science fiction end up on the horizon of about the next two years. So I asked Amodei on the show to share what he sees in the near future. What breakthroughs are around the corner? What worries him the most? And how are societies that struggle to adapt to change and governments that are slow to react to them supposed to prepare for the pace of change he predicts? What does that line on his graph mean for the rest of us? This episode contains strong language. Mentioned: Sam Altman on The Ezra Klein Show Demis Hassabis on The Ezra Klein Show On Bullshit by Harry G. Frankfurt “Measuring the Persuasiveness of Language Models” by Anthropic Book Recommendations: The Making of the Atomic Bomb by Richard Rhodes The Expanse (series) by James S.A. Corey The Guns of August by Barbara W. Tuchman Thoughts? Guest suggestions? Email us at [email protected]. You can find transcripts (posted midday) and more episodes of “The Ezra Klein Show” at nytimes.com/ezra-klein-podcast. Book recommendations from all our guests are listed at https://www.nytimes.com/article/ezra-klein-show-book-recs. This episode of “The Ezra Klein Show” was produced by Rollin Hu. Fact-checking by Michelle Harris. Our senior engineer is Jeff Geld. Our senior editor is Claire Gordon. The show’s production team also includes Annie Galvin, Kristin Lin and Aman Sahota. Original music by Isaac Jones. Audience strategy by Kristina Samulewski and Shannon Busta. The executive producer of New York Times Opinion Audio is Annie-Rose Strasser. Special thanks to Sonia Herrero.

Transcript

Click on a timestamp to play from that location

0:00.0	From New York Times opinion, this is the Ezra Klein Show. The really disorienting thing about talking to the people building AI is their
0:27.8	altered sense of time. You're sitting there discussing some world that feels
0:31.7	like weird sci-fi to even talk about and you ask
0:34.9	well when do you think this is going to happen and they say I don't know two years
0:40.1	behind those predictions are what are called the scaling laws.
0:43.4	And the scaling laws, and I want to say this so clearly, they're not laws.
0:47.8	They're observations, their predictions, they're based off of a few years, not a few hundred years or
0:54.2	thousand years of data. But what they say is that the more computer power and data you
0:59.2	feed into AI systems, the more powerful those systems get that the relationship is predictable and more
1:05.4	that the relationship is exponential. Human beings have trouble thinking in
1:10.6	exponentials. Think back to COVID when we all had to do it. If you have one case of
1:15.7	coronavirus and cases double every three days, then after 30 days you have about a thousand
1:20.1	cases. That growth rate feels modest, it's manageable. But then you go 30 days longer.
1:26.2	Now you have a million. Then you wait another 30 days. Now you have a billion.
1:31.6	That's the power of the exponential curve.
1:33.8	Growth feels normal for a while.
1:36.0	Then it gets out of control really, really quickly.
1:39.1	What the AI developers say is that the power of AI systems
1:42.1	is on this kind of curve, that it has been increasing exponentially,
1:45.8	their capabilities, and that as long as we keep feeding in more data and more computing power,
1:50.7	it will continue increasing exponentially. That is the scaling law
1:54.4	hypothesis and one of its main advocates is Dario Amadee. Amadee led the team at
	...

Please login to see the full transcript.

Previous episode | Next episode

Disclaimer: The podcast and artwork embedded on this page are from New York Times Opinion, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of New York Times Opinion and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.