pritam

Your next website visitor might not be human.

It might be an AI agent researching, comparing, and making decisions for your customer.

Today, we're introducing Agent Experience Optimization (AXO) to make websites adaptive for both humans and AI agents.

One website. Same URL. Different experiences.

Own the Experience.

#AXO #AgentExperience #AIAgents

I completely agree with @bchesky on this. Here's my theory.

There are jobs that are purely transactional in nature: comparing insurance costs, checking credit card eligibility, buying the same supplement I order every month, booking flights. These categories will be completely eaten by the consumer agent.

Then there are categories that are exploratory in nature, like figuring out a stay, which involves human judgement and a vibe check, or buying a recliner that suits my space. These will always be browsed by a human first. But yes, the number of sessions and the time spent in the app will go down, because the agent will narrow down the options for me.

So the point is, it's not one way for everything.

First principles: the world always moves towards better efficiency. Marketplaces came into existence because they solved for efficiency, so users didn't have to visit 5 different stores or hotel websites to navigate and transact. If consumer agents cut that pain completely and get the job done on behalf of users, my sense is the world will invariably move towards one less middleman, the marketplace. The inconvenience and lack of trust in handing your card to an agent will be solved in no time. Both websites and payment networks are already working to make the agent a first-class citizen.

I also very strongly believe agents accessing websites via a VM is just level 0 of the agentic timeline. It's where we first get to see how the world looks with consumer agents at work. Like everyone is arguing, this process isn't efficient yet. But we'll soon figure out the path to make it efficient by creating dedicated experiences for the agent.

@FireworksAI_HQ faced a massive security incident with potential business loss of $1 billion.

Founder-market fit is going to be the key moat going forward.