Zero-Shot Learning
00:23:07 · Nancy Wang, Adarsh Hiremath
[00:23:07] S3: Right. [00:23:08] S2: Again, the way to correct for that is an eval investment that is representative of the task that that agent is doing. [00:23:13] S3: And so that actually brings us to, I think, a fundamental question that, especially as a security company, we think about a lot, which is trust and the ability to handle edge cases. Because oftentimes, let's say in security, the difference between a false positive and a false negative is potentially your company ads have been headlines. So, the question there is, how do you think about last mile coverage? Do you go for more, let's say, you know, broad set, or do you wanna go for confidence on a very narrow slice of the problem? [00:23:41] S2: I typically say broader. The good part is that the models have reached a level of intelligence where they, if they have all the right scaffolding, can generalize. But you won't know unless you have like a broad sample of the problem, and you put that in an eval set. On the topic of 1Password and trust and safety, one of the problem spaces that I'm also really excited by is just authentication for agents. Right? Mhmm. Humans need this to Yep. To to make sure their activities are secure. Agents need this as well. I think there's a very strong case to be made that the addressable market of agents taking actions that need to be gated in the appropriate way is gonna be larger than humans in the next couple years, if it isn't already. So I'm I'm very, very excited about that, and what that means for 1Password and other companies focusing on security.
00:23:41 · Adarsh Hiremath, Dev Tagare, Nancy Wang
[00:23:41] S2: I typically say broader. The good part is that the models have reached a level of intelligence where they, if they have all the right scaffolding, can generalize. But you won't know unless you have like a broad sample of the problem, and you put that in an eval set. On the topic of 1Password and trust and safety, one of the problem spaces that I'm also really excited by is just authentication for agents. Right? Mhmm. Humans need this to Yep. To to make sure their activities are secure. Agents need this as well. I think there's a very strong case to be made that the addressable market of agents taking actions that need to be gated in the appropriate way is gonna be larger than humans in the next couple years, if it isn't already. So I'm I'm very, very excited about that, and what that means for 1Password and other companies focusing on security. [00:24:21] S1: That's basically definitely blooming. [00:24:23] S3: I would say, appreciate the call out. I mean, especially with agent swarms, I mean, you're already applying a multiplier effect on the human to agent ratio, and especially agents doing, you know, real work out there and taking real decisions. [00:24:34] S1: Even, like, out in delegation, right, like, you're talking to a third party SaaS from, like, any one p, need a mechanism and you need coordination across. So, yeah, that that space is definitely booming a lot. Related question to this, slightly different from the security space, Adarsh, the context of an enterprise evolves over time. [00:24:52] S3: Mhmm.
00:24:52 · Nancy Wang, Dev Tagare
[00:24:52] S3: Mhmm. [00:24:53] S1: Right? Like, today, I may be doing... Let's let's consider I'm a food delivery app. Tomorrow, I could do groceries. Day after, I could do something else. You know, in the future, I could do drone delivery, something like that. Right? So the context for an enterprise evolves. How do you keep this context evolution in perspective as you develop? And a related question to that is, in a more simpler way, how do you solve for data drift?
00:25:22 · Adarsh Hiremath
[00:25:22] S2: Maybe just a broader point is that people often view these investments in evals or whatever your infrastructure is as a one time investment. I think that that is extremely far from the truth. These are live investments. They're continuous investments. A good analogy is if you are shipping a feature, you would wanna write unit tests for that feature, assuming it's a net new feature. In the same way, if you are doing something with an agent, you would want an eval set that covers the scope of that agent and evolves over time. A concrete example of this in the customer support context, just because it's one that we're all familiar with, is let's say you have an agent that is trained to respond to inquiries about the product. Right? But then a new feature is released. Right? Like, there needs to be some steering and updating of the skills file to make sure that at the first time the agent fails on a query like that, there's automatically feedback that is inserted into the agent to address that new use case. And that is not something that's just true for customer support. It's also true for agents in in all contexts, their context. Right? Like, companies are idiosyncratic. They evolve quickly. They, add a bunch of new context. So that mechanism, that that you use to distill the company context into the agent and do it at really, really high velocity, keep the feedback loop short, is is a very, very, important one to solve.
00:26:35 · Dev Tagare, Adarsh Hiremath, Nancy Wang
[00:26:35] S1: So along the same, acts as other, Shah, if you think about, you know, side effects that an agent would have in interactions with the model, or the model is instructing something malicious or funny back to the agent, either directly or with some other feedback loops from other agents. How do you think about adding a layer of security in the middle, in which you can enforce some policies, you can, you know, evolve the intelligence of the agent alongside the model. What's your take on that? [00:27:03] S2: Yeah. So, you know, the first thing I wanna call out is that it's really, really hard to roll out agents to production in in part because of this issue. Right? And it's supported by the fact that the initial adoption of these agents has largely been for personal assistant use cases where the human is always steering the agent, and if the agent fails, it's on the order of magnitude of failing for one person as opposed to maybe a million people [00:27:24] S3: Yeah.
00:27:24 · Nancy Wang, Adarsh Hiremath, Dev Tagare
[00:27:24] S3: Yeah. [00:27:24] S2: That would be relevant to to a company. So just wanna say that that that's a really, really hard problem that a lot of people are trying to solve. In terms of how we actually steer the agents for a production context that has a larger blast radius, there's several components. I would say that there's just a natural progression of what you wanna do as you're in agent development life cycle. Right? The first one is you just try the agent in like a personal assistant context. Does it feel like the agent can do the task? Right? Then the next thing that you wanna do is then connect it up to some tools in a sandbox environment without the real data, validate that it can do the task again. Then you wanna roll out a set of evals, you know, live or not live evals to more methodically see if it could do the task. Then you wanna connect it all together in a sandbox environment that has real users, which means live eval connected to to real tools, but not real data. The agent is actually usable. And then you wanna do a phased rollout in a production context that has all of this infrastructure as well. So the key is not jumping the gun and following the whole process. [00:28:20] S1: Perfect. And like you throw an HITL somewhere in the loop with and without production. Right. That way, you have some control over what could happen, what should happen, so on and so forth.