ARC-AGI, every mention
21 scenes, the whole family · ← back to ARC-AGI
tap a year for its mentions
every year anyone François Chollet 25Garry Tan 12Diana Hu 8Ian Fisher 2
Verbatim, from the transcripts: the passages where ARC-AGI comes up
Boris Cherny: We Cut 80% of Claude Code’s Prompt · Y Combinator
World Models, JEPA And The Path To Sample-Efficient RL · Y Combinator
- ▶ 1:09 unnamed speaker And so if I imagine I have a new tasks like RKGI, for example, I think like really Francois Chalet has been on the forefront of this thinking, uh, and talking about intelligence as a rate of skill acquisition versus skill acquisition. 2 times in the scene
François Chollet: Why Scaling Alone Isn’t Enough for AGI · Y Combinator
- ▶ 0:04 François Chollet Around the time, uh, that we're going to be releasing like maybe arc six or arc seven.
- ▶ 0:31 Garry Tan Today, we're lucky to be joined by Francois Chollet, founder of the ARK Prize, a global competition to solve the ARK AGI benchmark. 2 times in the scene
- ▶ 9:42 Diana Hu I think I really like how you define intelligence and how to measure it, which brings to the question of, uh, also sharing, having you share the history of, uh, RKGI.
- ▶ 12:36 unnamed speaker To Diana's question, do you want to talk about how you actually designed Arc AGI and why it's a good barometer of that? 3 times in the scene
- ▶ 17:25 Garry Tan You're just ahead of your time, clearly, because we're not on Arc AGI one anymore, and then two is reaching saturation, and then three is out now. 7 times in the scene
- ▶ 17:36 Diana Hu And I think the cool thing about Arc AGI, it has been a very good barometer for the industry of the big changes that happen, because V one was not working at all for a long time until 2025. 4 times in the scene
- ▶ 24:20 Diana Hu And they did that basically for arc V two, which when they released it, 2 times in the scene
- ▶ 26:09 François Chollet So if you look at V-one, V-two, it was really focusing on your ability to produce like causal models of a pattern that was just given to you, like the data was given to you. 3 times in the scene
- ▶ 31:29 François Chollet You must, you know, in, in, in an IQ test, like problem, like, you know, what arc one and two is, the data that you must model is provided to you.
- ▶ 44:11 François Chollet Um, the, the point of the arc AGI benchmark series is not to say that, well, you know, here's this test.
The Powerful Alternative To Fine-Tuning · Y Combinator
- ▶ 5:07 unnamed speaker Like, I remember when you first came out with your paper in December of last year, uh, you shot to the top of Arc AGI V-II. 2 times in the scene
- ▶ 12:20 Ian Fisher It, it was pretty interesting to look at the, um, prompt outputs in particular, I'd say, for ArcGi in that, uh, you know, I think you can read those and say, well, that's not what a human would have written.
- ▶ 15:46 Ian Fisher And, you know, we view the Arc AGI results and the humanities last exam results as showing kind of two different, uh, capabilities that we have.
The Truth About The AI Bubble · Y Combinator
- ▶ 25:33 Garry Tan Um, the ARK AGI prize is, uh, you know, gonna come in and do the winter 26 batch as a non-profit. 2 times in the scene
- ▶ 25:39 Garry Tan The funny thing about that is like, yeah, maybe there's, um, a team right now that is climbing the leaderboard of ARK AGI and they're gonna accelerate this thing again.
François Chollet: How We Get To AGI · Y Combinator
- ▶ 2:04 François Chollet It's called the abstraction reasoning corpus. 5 times in the scene
- ▶ 11:49 François Chollet My first attempt at creating a way to measure intelligence in AI systems was the ArcGi benchmark. 10 times in the scene
- ▶ 20:18 François Chollet What's it going to take to solve Arc-II, and we are still very far from it today, then solve Arc-III, and we're even further away from that. 3 times in the scene
- ▶ 34:29 François Chollet And to test our progress, our first milestone is going to be to solve RKGI using a system that starts out knowing nothing at all about RKGI.