The Open Source AI Question (Robert Wright & Nathan Labenz)

Robert Wright's Nonzero

Nonzero

News & Politics, Society & Culture, Philosophy

4.7 • 618 Ratings

🗓️ 15 March 2024

⏱️ 62 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

This is a free preview of a paid episode. To hear more, visit www.nonzero.org

0:33 Nathan’s role as “AI scout” 4:07 What "open source" in AI really means 9:45 Why the secrecy around AI training data? 17:19 What’s actually “open” about OpenAI? 32:56 The case for and against open source AI 42:35 Is an open source crackdown coming?52:25 What Waymark, Nathan’s company, does with AI

Robert Wright (Nonzero, The Evolution of God, Why Buddhism Is True) and Nathan Labenz (Cognitive Revolution, Waymark AI). Recorded March 13, 2024.

Twitter: https://twitter.com/NonzeroPods

The Cognitive Revolution podcast: https://www.cognitiverevolution.ai/ Waymark AI: https://waymark.com/

Overtime titles:

0:00 Bob glimpses a Matrix-like future 4:34 Could (and should) we use AI to resolve human disputes? 9:47 Where Nathan falls on the AI doomer spectrum 22:50 Understanding how AI understands things 32:22 Is AI alignment overblown?

Transcript

Click on a timestamp to play from that location

0:00.0	You're listening to Robert Wright's Non-Zero Podcast.
0:33.0	Hi, Nathan.
0:34.6	Hi, Bob.
0:35.5	How are you?
0:36.7	Great.
0:37.2	How are you? I can't complain. Let me introduce
0:39.6	this. I'm Robert Wright, publisher of the Non-Zero Newsletter. This is the Non-Zero podcast. You're Nathan Levinz,
0:45.6	a host of the Cognitive Revolution podcast, a highly regarded podcast about AI. I'm a regular listener.
0:53.5	You're also founder of Waymark, which is a
0:56.2	company that I guess I'd say helps mainly small companies with a limited ad budget make
1:02.2	video ads and it harnesses AI to do that in some kind of interesting ways. I actually went to the
1:07.6	site and went through the tutorial. It's kind of shockingly simple in principle.
1:12.7	And maybe we'll have time to talk about that.
1:15.7	But one thing we're definitely going to talk about is artificial intelligence, that's for sure.
1:21.7	You know, how powerful you think it's going to get, what risks you think it poses,
1:30.2	maybe what should be done about it.
1:35.5	I wanted to start out talking about the open source question, whether open source, whether powerful open source AI models are dangerous, if not like the current generation, maybe the next generation.
1:41.5	That's been in the news for a couple of reasons. One is it figures
1:46.0	in the question of whether Open AI betrayed its founding mission, a question that has been
1:50.7	raised by Elon Musk's lawsuit. I think there may be some misunderstandings about the extent to which
1:56.1	open source per se explicitly figured in the founding mission of open AI, but we'll get into that.
2:02.6	Also, this State Department commissioned report, not to be confused with the State Department
	...

Please login to see the full transcript.

Previous episode | Next episode

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

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