Agentic AI by 2035? The Future That Could Change Everything

Category: AI Trends

Tags: agenticAIautomationautonomyworkflows

Entities: Adept AIAgentic AIAGIAI agentsAlex WongAnthropicCognitionDario AmenDavid LuanDevonScale AIScott Woo

Transcript

00:00

[Music] The impact of agentic AI on the next decade may exceed that of the industrial revolution and the rise of agentic AI could transform the next decade more profoundly than the internet and the smartphone combined. We've always

00:15

thought about it as a system that has the ability to exhibit all the cognitive capabilities we have as humans. And the reason that's important, the human mind is the only existence proof we have maybe in the universe that general intelligence is possible.

So if you want to claim sort of general intelligence

00:31

AGI then you need to show that it it generalizes to all these domains and that's where the idea of agentic AI comes in systems that don't just respond but act reason and plan across domains like humans do we sort of name this the the calm before the storm uh because you know I think to

00:48

a lot of people in the AI industry itself it feels sort of very clear that we're basically on the precipice of AGI I [Music] one of the most exciting technology trends right now is a rise of agentic

01:03

workflows in AI or sometimes we call them AI agents. So with agentic workflows, we end up taking a much more iterative approach to deriving your results.

And I've seen in many applications this type of iteratively agentic workflow does much better than forcing the eye to grow from start to finish. Modality, memory, these are

01:20

things that we're working on and making good progress. But it's missing like planning uh and actions, the ability to achieve a goal that you might set a system.

So it has to be able to act in the world, take actions. Um if the objective is pretty complicated or

01:35

abstract, it has to have the ability to break that down into a set of steps that are achievable, a set of interim goals, interim plans, uh and then that to build up. [Music]

01:50

Agents despite being the thing that everybody talks about uh are not here and are not reliable. We see this very strong parallel between these autonomous AI agents and uh this L1 through L5 progression that we saw in autonomous vehicles.

L1 is maybe just a chatbot. L2

02:07

is a assistant that you can ask for help on various things. And then we're sort of charting this interesting path I would say of L3 to L5 autonomy.

L3 agents that you can use for discrete portions of workflows. So you can start using them as things that you can basically start relying on.

L4 of

02:23

autonomous agents that um when they need human help, they'll like ask you. And then L5 is can sort of all just go on vacation and uh hang out.

Across the AI industry, a new shift is underway. And instead of answering questions or generating text, AI systems

02:40

are starting to act as agents. They're taking initiative using tools and making decisions.

And the people leading this change are some of the most influential voices in the field. AGI, AGI, AI, AGI, AGI, AGI, artificial general intelligence is a system that can exhibit all the

02:56

cognitive capabilities humans can. In their own words, here's how they describe the rise of Agentic AI.

Agentic AI isn't just theory anymore, and the leaders building it are already seeing it reshape the world. Alex Wong, CEO of Scale AI, put it bluntly.

You're going to see agents embed across military,

03:11

enterprise, logistics, everywhere. And Dario Amen, CEO of Anthropic, described how his team approaches this new reality.

You see, we design agents with constitutional principles that guide our autonomy. David Luan, CEO of Adept AI, explained how agents behave in the

03:26

workplace. The agent understands and operates the software just like a human teammate.

And Scott Woo, CEO of Cognition and creator of Devon, showed us just how far autonomy has come. You give it a task, it writes code, debugs, and pushes to GitHub all without you.

03:43

We know that people want AI that is smarter, more personal, more customizable, can do more on your behalf. Eventually, you'll just ask a computer for what you need and it'll do all of these tasks for you.

These capabilities are often talked in the AI field about as agents. The upsides of

04:00

this are going to be tremendous. We're at a tipping point.

Uh this is clearly the age of co-pilots. uh from digital natives like Airbnb or Dolingo to uh to Shopify as well as the world's largest companies uh whether it's BT or Bayer or Densu goody lumen

04:18

companies in every industry who are building their own co-pilots uh from Else in finance to epic in healthcare uh to Rockwell automation in manufacturing and Seammens in manufacturing it's fantastic to see people deploy their own

04:34

co-pilots I think AI is is one of the biggest threats and here we are for the first time really in human history with something that's going to be far more intelligent than us. It's not clear to me we can actually control such a thing, but I think we can

04:51

aspire to guide it in a direction that's beneficial to humanity. But I do think it's one of the existential risks that uh we face and it's it's potentially the most pressing one.