How Agentic Service Discovery is changing the interactions between customers, brands and products

Organizations like Google (Gemini), OpenAI (ChatGPT) and Anthropic (Claude) are improving the AI models in a race-like fashion, competing to have the best, fastest and most capable models. Parts of the development have brought us the concepts of AI agents and Agentic AI. AI agents are already commonly used for many tasks, ranging from administrative tasks like organizing company files and data, writing meeting notes to coding websites and deriving insights from complex data sets. AI agents are basically becoming small experts where people can get help with more or less anything and the power of the AI community is already building agents for all kinds of things. While this is great, these agents must still be configured to your personal accounts and device, before you can get them to help you.

This means that if the coolest sports fashion brand in the world wants potential customers to use their AI agent that can suggest the best running shoes for your personal needs, the customer must configure the AI agent before being able to interact. In other words, the entry barriers are not intuitive and can be too technical to many customers. Furthermore the brand must market the AI agent to create awareness, which increases the risk of adoption.

Think of this as the first phase of how customers and brands can interact through AI. The adoption will likely only come from technical businesses and customers, while others will continue to have conversational interactions with their go-to AI chat agent only.

The issue is the lack of standards in the AI industry. But this has already started to change.

Any great technology needs standardization to reach its potential of mass adoption. Where the AI chatbots have already gained massive popularity by using standards like the internet and web browser to interact with the users, AI agents are still at an early stage in terms of achieving the same level of adoption.

That said, initiatives are already evolving to standardise and bridge the pieces together for a more unified AI ecosystem. Initiatives like MCP (Model Context Protocol) and A2A (Agents2Agent) are examples of how to address some of the challenges, like making the interaction with AI agents more intuitive for everyone, brands and customers.
So what is exactly missing? Agent Discoverability - the infrastructure allowing AI agents to browse repositories of other AI agents and directly interact with them on behalf of the user. Yes, it’s a bit of a mouthful.

This sounds like what is already possible today, but the standards are not entirely there yet. As mentioned before, users can only interact with AI agents that they have actively configured to use. This contradicts the ability to browse any public AI agent and use it for responding to the prompt of the user.

There are multiple reasons for the challenges to achieve something like Agent Discoverability, like data security and authentication. But for now we want to focus on how an AI maturing ecosystem will impact the interaction between customers and brands.

If you have used a go-to AI chat bot, you can maybe relate to how easy and intuitive it is to find information, get suggestions and advice or rewrite your emails. Now try putting this in the context of a consumer's lens. “What are the best running shoes on the market right now?” - could be a prompt to your designated ChatGPT, Gemini or Claude chat agent. With a matured AI ecosystem where AI agents can browse and discover other agents (Agent Discoverability), the response to your prompt could be interaction with AI Agents from one or more brands instead of a simple written answer. Depending on the AI Agent the brands have made you could be asked a series of follow up questions, to eventually find the best running shoe for you.

But it doesn’t stop here. With the reasoning capabilities of the AI models, the fashion brand can have multiple Agents for different scenarios. Now a sales AI Agent takes over and makes everything ready for you; the shoes are added to the basket, it’s asking you to validate your shipping details (which I has from your profile) and in the end ask you to confirm the purchase - all of this happening in an instance without you having to leave the AI prompt. This is the light version of how the future AI ecosystem will impact the relationship between customers and brands.

So, AI can soon complete transactions - It doesn’t really sound overwhelming revolutionary when thinking about what AI in general is capable of. Here’s the thing; with a standardized, interconnected AI ecosystem, brands can have AI Agents ready to respond to any kind of request coming in from prospects, current customers, returning customers, customers having questions about their products, prospect comparing services across brands and so on, and use AI and LLMs to give the user a relevant experience.

Even if the AI revolution seems like science fiction, you can ask yourself the following:

• Is it more intuitive for consumers to chat with AI when researching products, than browsing through search engine results and websites?
• Will the younger and next generations grow up with AI as their primary route for finding information, products and anything else that AI Agents?

Fugentic is following this development closely and will be to offer brands the necessary tools and integrations in an easy and simple way. Sign up for our updates and get free access to our evolving platform.