Google pairs model access limits with runtime safeguards

Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

1. 1:49

   You were mentioning the compute constraints. It's interesting that even a company like Google has those for a new model like this. I mean, Argonne's a large model. It's obviously a very expensive model to serve. Can you walk me through the decision of how you have to now gate this? I believe right now it's accessible through the API, AI Pro and Ultra subscribers, obviously not to the widest possible user base yet.

2. 2:10

   And I imagine that's challenging for a model of this size. So how do you think through that, gating that and releasing it? One of the things we've been very good at, if you go back to the 3.0 models, we can distill of these models and get versions which are nearly as capable, but much more efficient. So we are working on those models as well, which will help us give wide access to as many people as possible.

3. 2:36

   So those efforts are underway. We also, with this model, we are definitely at a moment as we are working through Frontier. you want to really think about safety of these models. So we've invested a lot more effort, both in terms of how we tested it externally through our Fairwind program, the U.S. government giving early access to cyber researchers, internally, widely testing it. And we've obviously brought it out externally.

4. 3:06

   You're right, it's a large model. We are serving it pretty fast given the size of the model. So I'm excited what we are doing, but we'll continuously expand access from now on. And so you saw at the launch time, we had made it available to all ultra tiers, pro tier in the Gemini app. We followed it with pro tier in the anti-gravity app. And we are working on distilled versions, which we can serve to more people with the 4.0 series.

5. 3:38

   Is this the carbon model that there's rumors about? At any given moment in time, we have internal versions of our Frontier models, and so some of these are code names of those versions. That's different from us getting a more efficient version of Argon available more broadly. So more efficient version coming. of testing it for safety i'm curious to hear you talk about that it's it's interesting to think about how the private frontier labs open an anthropic talk about the security and privacy of their new models versus a google or a meta or you know a larger hyperscaler it sounds like you guys are being very methodical with the safety testing for this one and it's maybe taken longer than previous models? Is that fair to say? Definitely. I think with the agentic capabilities and its ability to code so well, particularly when you're deploying it, you know, we are serving

6. 4:37

   many enterprise customers at scale. You want to make sure, so for example, the work we've done, obviously the work you do to align these models well, but more importantly, both the compute and the systems you build to monitor them while they are working, right? And that can be compute intensive too, but that real-time monitoring as these models are deployed, I think ends up being critical so that you can, you know, you get as early signals as possible if something is not working as intended and you can intervene.

7. 5:10

   So we have taken time to build those systems, the sandboxing you would need to do, the real-time monitoring you would need to do, and with that, also for customers, giving them the protections they need so that you exactly know where it's running, what access it has, all the telemetry you need, all of that. This interview is presented by Hyperagent.