AI:AM Highlights: Astra as AGI, OpenAI's Pause, Mythos @ Mozilla & Human Agency vs Technocapitalism

Nathan Labenz (0:02)

My co-host, Prakash Narayanan, after a weekend working with GPT-6 Astra.

Prakash Narayanan (0:08)

People are going to use this thing. Token spend is going to increase dramatically. I think a lot of people are going to be using it all the time.

It is AGI. It is that kind of cleared the hurdle of AGI. It will do things better than most people you can hire and train.

Nathan Labenz (0:29)

Welcome to the AI and the AM weekly highlights. This is Nathan, using my cloned voice to introduce clips from our three live shows this week. Let us know what worked and what did not.

Part 1, The Astra Weekend, Tuesday, September 8th. Here is what Prakash had been building.

Prakash Narayanan (0:52)

I spent the entire weekend using Astra.

I was running three to four agents continuously, and they were good. Astra is very, very good in the sense that it started to tackle those annoying problems which had been in the codebase. As you know, we built the studio by ourselves.

It started to tackle some of the long-standing issues in the codebase, which had been kind of annoying and bugging me, and it started to resolve those issues. It is very, very good. I would say it is finally at the point where, if you care about the quality of the work, you can still hand it off to Astra. But you still need to do a little bit of talking, but you can hand it off to Astra, and you can get some results.

And the computer use is good. The other thing that was failing really badly, I think, before is computer use. Computer use on GPT 5.6 would sometimes take a very, very long time to click around and do a bunch of stuff, and computer use finally works properly in the time frame that you give it. So it's cleared the hurdle. It's cleared the hurdle of genuine usefulness at this point. And you can start to give it more advanced tasks. So this is a guy called Skalsky. So he trained models to identify players on the basketball court. He hand labeled 12,000 individual images with who the players were, referee or this player, that team, etc. And he hand labeled 12,000 images.

And now, Astra can just do it, like Astra just doesn't. This is a task that a human being will never do again. Like there just isn't any point. You can't even pay someone to do it because if you paid someone to do it, they would use Astra to do it and then like pass you back the results. Like it's done, like a human will never do this task again.

Nathan Labenz (3:22)

I had been watching how Astra keeps working through long tasks and how it uses notes to stay on track.

How is it that these new models are so persistent, right? How is it that they can come up with such elaborate chaining togethers of all these different exploits to finally accomplish a goal that, you know, if we had to do so many things, we would just give up, right? And most models historically weren't able to do it. It seems that they have a new way of handling history, which, like many brilliant insights, seems pretty obvious in retrospect. But nevertheless, it is new, and maybe Anthrocynics isn't doing this and hasn't said it, but what I understand Astra is now doing is, instead of compacting its million tokens into a summary and then essentially starting a new context window with that summary, but losing all the detail that was summarized away in that compaction process, now there is a long-lived notes file that the model can update whenever it needs to. And this kind of follows it forward in time, regardless of how many tokens it's laid down. And then it has the ability to go back and search through its own session history. So now, even though you only have maybe still the same million-token context window, a million-token is all it can handle in one shot, fully attending everything to everything, it has enough via the notes and the ability to go back and search and see what's done before to pretty effectively manage, it seems like at least 10 times that much, context in single rollouts.

How do you measure what these models can do now? Prakash started with the chart from METR, the research group that tracks how long a task AI agents can complete.

Prakash Narayanan (5:26)

You know, one of the people online, Ethan Malik, who is a professor who tests a lot of models, he posted the METR, the famous METR hours of work chart. There hasn't been an update for a while now.