How AI is saving billions of years of human research time | Max Jaderberg

TED Talks Daily

TED

Creativity, Ted Podcast, Ted Talks Daily, Business, Design, Inspiration, Society & Culture, Science, Technology, Education, Tech Demo, Ted Talks, Ted, Entertainment, Tedtalks

4.1 • 11.9K Ratings

🗓️ 2 December 2024

⏱️ 19 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

Can AI compress the yearslong research time of a PhD into seconds? Research scientist Max Jaderberg explores how “AI analogs” simulate real-world lab work with staggering speed and scale, unlocking new insights on protein folding and drug discovery. Drawing on his experience working on Isomorphic Labs' and Google DeepMind's AlphaFold 3 — an AI model for predicting the structure of molecules — Jaderberg explains how this new technology frees up researchers' time and resources to better understand the real, messy world and tackle the next frontiers of science, medicine and more.

Transcript

Click on a timestamp to play from that location

0:00.0	TED Audio Collective.
0:09.0	You're listening to TED Talks Daily, where we bring you new ideas to spark your curiosity every day.
0:15.1	I'm your host, Elise Hugh.
0:16.9	Today, a breakthrough in science thanks to the neural networks of AI that is saving years of human research time.
0:25.1	In his 2024 talk, AI researcher Max Yoderberg makes the case for something called AI analogs.
0:32.1	And he explains what these advances can do to allow for more experimentation, understanding, and new knowledge.
0:39.5	Coming up after the break.
0:44.9	TED Talks Daily is sponsored by Stripe.
0:47.7	Stripe helps many of the world's most influential businesses grow their revenue and build a more profitable business.
0:53.9	Whether it's hurts making a
0:55.1	checkout a smooth ride for their customers, OpenAI answering unprecedented demand, or PGA
1:00.9	chipping away at back office inefficiency. Stripe's financial infrastructure platform helps
1:06.0	companies achieve ambitious goals. No matter what success looks like for your business, Stripe helps ensure the
1:12.2	complexity of financial systems doesn't get in your way. Learn more at stripe.com. And now our TED Talk
1:21.2	of the Day. So while I go now, I did a PhD, and I actually thought it'd be quite easy to do research.
1:29.3	Turns out it was really hard.
1:31.3	My PhD was spent coding up neural network layers
1:36.3	and writing kuda kernels, very much computer-based science.
1:41.3	And at that time, I had a friend who worked in a lab doing real messy science.
1:48.0	He was trying to work out the structure of proteins experimentally.
1:53.0	And this is a really difficult thing to do.
1:57.0	It can take a whole PhD's worth of work just to work out the structure of a single new protein system.
	...

Please login to see the full transcript.

Previous episode | Next episode

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

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