Autonomous Agents

“If you have too many rules, they always conflict. It's almost unsolvable. There's some number, I don't know, there's some number of Dunbar, there's a Dunbar number for rules where you get out to 40, 50, 60, 70 gates on a process. Poor instinct and GrokBot can'…”

“And as co-work becomes, you know, more successful and more stable and more broadly used, I would be really concerned about SaaS companies that don't have some kind of system of record or some kind of AI strategy in place to succeed. Because the co-work era, it…”

“And you can like optimize, not just for getting the work done, but for making it easy for your reviewer to understand the decisions that you made. And that's obviously very true in software engineering, and it's actually true in, I think, most professions and …”

“We've got devs in India who all of a sudden are driving 10 X the code commits that they used to. They're using autonomous agents all over the place.”

“We had a cloud code, which became like the really first widely used autonomous agent.”

“And so I think that today autonomous has by far the fastest growing category and today typically comes without any controls.”

“The place we're going is one where you as A person doing work that you are the overseer. You are the CEO of almost this autonomous corporation, or, you know, of this fleet of agents perhaps is more, is, is the way to say it, and that they are operating accordi…”

“I have Absolutely no doubt that everybody in the world is gonna have at least, you know, an agent like this, if not an entire family of agents, and we're gonna be living in a world where I think it's almost inevitable now that this is the way people are gonna …”

“The second cataclysm of AI is gonna be, you're just gonna, like, have all these agents running around doing crazy shit, and including things they have no idea what they're doing, and they'll check the APIs, be like, yeah, that thing works, and it sort of will,…”

“These are like the, some of the most probably, you know, boring problems for 98% of people on, on the internet, but they will be the problems that are the difference between, can you actually have autonomous agents in an enterprise context that are not leaking…”

“And in this case, it's autonomous agents. Autonomous agents is, is in my opinion, what the tsunami is about. Not just AI in general.”

“So if I look at real true AI startups like E to B or eventual or Lotus AI get dynasty. These are all native AI companies that are up and running Ovens, another great one. What's happening with them is they're all using open claw, nano claw, or one of their own…”

“I think what's going to happen with the autonomous agents is ultimately they're going to have an orchestration layer where they can have multiple different LLMs, you know, that they can orchestrate on and triage workflows.”

“And my guess is this will lead to the rise of open source models more, more proficiently. And that it's going to lead to the rise of ASICS chips, because what's going to happen with ASICS is you're going to put the model on the chip. You know, ASICS is going t…”

“Autonomous agents, and it may take a year or longer for most enterprises to actually enable and get committed to autonomous agents, but they're going to write software faster, cheaper than any human being on the planet.”

“But if you're not making your software for autonomous agents today, you're going to be challenged in the future. Maybe it's six months, maybe a year, maybe 18 months, but you're going to be severely challenged if you still think human beings are going to buy y…”

“There's no question that anyone that inputs data into a computer that does scheduling that, you know, like an executive assistant or people doing marketing, those jobs are probably ultimately going to be better done by autonomous agents, white collar jobs.”

“First of all, not to the people with the job. First of all, to the next person in line, the next junior developer you want to hire, the next executive assistant, the next marketing person. So the first thing you see is people stopping or slowing down Hiring of…”

“I think the first companies that are going to go with this are going to be small, medium businesses, because they're the ones that, wow, one secretary makes a huge difference in their, in, in how they, you know, two, three, four person company, how they handle…”

“And so the one thing that the autonomous agents aren't going to have is intuition. Not anytime soon.”

“Language is, is a very subpar Descriptor of the world. It's just limited. So for autonomous agents, we need a new system where we can visually recognize the world to be able to have an objective view of reality for the autonomous agent.”

“Absolutely. Absolutely. The best time because you're in a sea change. Humans are no longer going to be the decision makers. About software, it's going to be autonomous agents.”

“One of the things that we are experimenting with this mission control in GitHub Copilot is that, right, which is, well, the idea is you go have five, six different branches in which you fire off all these autonomous agents, they all do their work, they come ba…”

“The concern, especially as you move from generative to agentic AI, where you have autonomous agents moving through your systems doing stuff without checking back in with a human, I think it creates risks that we haven't got our hands around yet.”

“No, look, I think the most vulnerable people are where there is poor loyalty to the UI.”

“Working with valuable LLMs and so on, I would say, you know, it's less Iron Man robots and more Iron Man suits that you want to build. It's less like building flashy demos of autonomous agents and more building partial autonomy products”

“So I do not believe that autonomous agent would be a runtime phenomena. However, there would be a build time or compile time phenomena.”

“And I think the reason why we care so much about reasoning is because I think that's the path that we get reliable agents through.”

“METER put out a study of the agentic work that can be done you know, by a wide range of benchmarks and dated it, ran it all the way back to 2019, and basically came out with the idea that the agent horizon for the 50% capability, like, 50 percentile capability…”

“The cool thing is, though, is that, like, you know, it's a nines of reliability scaling problem, that as you, like, kind of, like, push these nines out, and as you, like, make them more reliable in many more contexts at some point they are just doing, like, a …”

“This could possibly change overnight, but it's still near the floor. I think they're still extremely defective as agents.”

“Three is agents ability to go off and do these longer-term tasks You know, maybe like multiple interactions with an environment, asking people for help when they need it, working together, all of that. And I think we're gonna get there faster than people expec…”

“Critically, last line here on the slide, the workflows with AIs Currently produce better content than autonomous agents and better content than humans alone. Unless you're ready to pay, like, 10,000 dollars per post.”

“So this is why, in my opinion, the first We will get to autonomous agents, but by going through, first, we will have to have this digital assistants that work under people's supervision.”

“And then we've got another big wave coming very, very close. Which is going to be autonomous and agents. And then of course we'll get to AGI.”

“That's something that seems to be missing from these autonomous agents right now. Everything is explicit reasoning, and at some point that might not scale or that might become brittle.”

“The basic idea is to orchestrate LMs in this like iterative loop towards some high level goal. Where they're doing planning, and if memory, and prioritization, reflection, and so you're not necessarily changing the architecture of the LLM itself, but this orch…”