Agentic adoption in UI/UX.

At Atmn, we define UI-embedded AI agents as autonomous decision making units within any user interface.

We estimate human users to consistently use a maximum of 100 agents at any given moment in time.

This assumption can be made based on several postulates like Dunbar's Number, etc. Thus limiting the average number of agents used by individual users to be closer to the average number of applications used by an individual user today i.e. ~10 agents daily or ~100 agents monthly.

We thus define 3 waves of UI-embedded AI agents adoption for businesses.

In the first wave of adoption of UI-embedded agents, we expect these agents to augment & replace app features, overhauling the user experience.

Several read-only & write-only app features which require human decision making will compress into singular agentic features.

This compression frees up the digital real estate on interfaces, allowing application developers to expand functionality.

Once the number of agentic features, increases by an order of magnitude, it marks the advent of wave 2, where users will work more with coordination or conductor agents.

Coordination or conductor agents are embedded-UI agents which interact with 10s, 100s & 1000s of other agents on behalf of a human user.

Number of coordination agents increases exponentially in Wave 2 until the applications become a collection of agents coordinating among themselves to deliver automated experiences to users.

At this point, agents will break out of the application layer, marking the advent of Wave 3.

In Wave 3, applications will be replaced with brand agents which take care of specific needs of a human user.

Human users will choose between competing brand agents, based on agent personas, trust, etc.

It is to be emphasized that Porter’s competitive strategies will be relevant across all waves of agentic evolution. Cost leadership or differentiation will always be the preferred paths to market dominance.

All businesses will find themselves traversing through these waves within the next 10 years.

Business entities will take up these evolutionary journeys, entering at any of these waves or through any intermediary stage.

While we overestimate the ability of incumbents to take up this evolutionary journey, those which struggle to take it timely will be replaced by the upstarts in their domains.

This is our attempt at understanding agentic adoption in UI/UX through an evolutionary model.

We expect this understanding to evolve further with continued advances in AI research & development. We will keep you posted so do keep yourself in the loop. 

-@AgrawalMayankG