What Is Agent AI And Why All The Excitement?

An unofficial definition and overview of the new hotness

If you hang out in A.I. circles, you’ve probably run into the term “Agent AI”.  Even if you don’t hang out in A.I. circles, you’ve probably still run into Agent AI.

And, you may have wondered, what exactly is Agent AI and why are the people that are excited about it so excited about it?

A Term Is Born

Anytime there is a new term that starts to get used in tech, one of the challenges is that nobody really knows what it means. People just start using it and repurposing it to mean what they need it to mean. 

For Agent AI, there wasn’t a committee somewhere that declared:  “Behold, we have a new thing, we call it AGENT AI!  Marvel at its magnificence! And, as of this day forward, we declare and define “Agent AI” thusly…” [insert definition here].

Nope, that’s not how it works at all.

Basically, somebody starts using the term because it’s a useful label for whatever it was they were thinking of, and then other people start using the term and over time, we as a community start to develop some sense for what it is.

A term is born.

But, it’s often useful to take a step back and at least try to define what a term means (with the understanding that others will critique it and make it better).

Totally Unofficial Definition Of Agent AI

So, here’s my totally unofficial, first pass at a definition:

Agent AI: Software that uses artificial intelligence to  pursue a specified goal. It accomplishes this by decomposing the goal into actionable tasks, monitoring its progress, and engaging with digital resources and other agents as necessary. 

Actually, it’s not my first pass. It’s like my 17th pass.

15 of those passes involved vacillating between whether to include the word autonomous in the definition. One could make the case that part of what makes agents special and exciting is that they can run autonomously with no human intervention. But, I decided to leave it out, because although that’s certainly exciting, I don’t think it’s a requirement for an agent to run completely autonomously. Based on the goal (and one’s tolerance for risk), it may be completely fine to have some human intervention and nudging in there.

An agent’s still an agent, no matter how small, and non-autonomous. (h/t to Dr. Seuss).

OK, so what makes an agent an agent and not just sparkling A.I. software from a certain region of Silicon Valley?

Short answer: Goals.

The Key Difference With Agent AI

Today, when we (as humans) interact with A.I. it’s usually through a conversational chat interface as popularized by OpenAI’s ChatGPT. You give ChatGPT a task by typing in a prompt, and it goes and does that task.

Examples:

  • write me a 500 word blog post about the impact of AI on CRM

  • List the top 10 cities in Italy including what they’re known for. Give that to me as JSON.

The fundamental thing that makes agents different is that instead of specifying a discrete task you instead specify a goal (what you’re looking to accomplish).  The agent that determines what tasks need to be completed in order to accomplish the goal. And, it’s smart enough to know how to break-down those tasks into sub-tasks and track its progress along the way.

Here’s an example of a high-order goal:

Launch a new online newsletter about Agent AI.

You can imagine this breaking down into something that looks like this:

  • Branding

    • Come up with name for the newsletter

    • Create logo for the newsletter

    • Figure out website domain

  • Tech Stack

    • Determine where to host the newsletter

      • Research available products

      • Gather ratings/reviews

      • Summarize pricing

      • Choose a platform to host

  • Writing

    • Write initial posts

      • Introductory post 

      • Flagship post

  • Launch

    • Make subscriptions available

    • Announce the newsletter

      • Available social channels

You get the idea. Each of the tasks above could likely be further decomposed into sub-tasks until you get to a level of granularity such that the task itself can be executed by the software. Or, instructions/guidelines can be written that can then be done by a human.

The big point here is that the we are specifying a goal or objective and letting the agent figure out how to get that goal accomplished.

Agents Taking On “Roles”

Let’s wrap up with why so many are excited about the potential (and perils) of agent AIs.

One can imagine building agents that have a specific set of goals based on a given “role”. Examples could be: designer writer, researcher, SEO, analyst, project manager, etc. This maps to actual roles that might exist in an organization.

Now imagine, having a team that is a mix of people and agents that can accomplish a goal more efficiently and more effectively than humans alone.

That is the promise of Agent AI.

Stay tuned for more, and thanks for subscribing.