Why companies are replacing generic AI with tools that know their users

The future of AI is not just agent; It is a deep customization.

Instead of simple recommender systems that correlate user behavior to identify patterns and apply them to individual workflows, large language models (LLMs) and AI agents can analyze users directly to create deeply personalized experiences.

It’s this kind of aggressive personalization that users are increasingly demanding, and the smartest companies that provide it (and soon) will win.

The goal is: “Don’t try to randomize or guess who I am. I’m telling you, this is what I care about,” explains Lijuan Qin, head of product at Zoom AI, in a new article. Beyond the pilot podcast.

How Zoom is incorporating personalization

Zoom is one company that has adapted to this trend: its generative assistant, AI Companion, goes beyond basic summary, smart recordings, and post-meeting actions to divergence of opinions and tracking user alignment.

Users can customize meeting summaries based on their specific interests and create specific templates for follow-up emails for different people (whether a salesperson or account executive). The AI ​​assistant can then automatically complete these documents after the call. Meanwhile, a custom dictionary in Zoom AI Studio can process unique business terminology and vocabulary for more relevant AI results, and a deep research mode can quickly deliver comprehensive analyzes based on “internal experience and external insights.”

Control is key here; the human can be “very specific (and) precise” in granting permissions to the agent, Qin explained. They have “very clear controls” over follow-up actions, such as: Can the agent automatically send emails to specific recipients? Or will it trigger a verification step when it recognizes that the transcripts contain sensitive information (as dictated by the user)?

Knowing that AI can go off the rails at times, human users can track agent behavior on Zoom, enable and disable features, and control access to data. This can help avoid inaccurate or skewed results.

“The most important thing is that we don’t assume that AI is smart enough to do everything right,” Qin emphasized.

Get the correct context

In this new era of agent AI, there is essentially a “land grab for context,” Sam Witteveen, co-founder of Red Dragon AI and host of Beyond the Pilot, explains in the podcast.

“Knowing your users is definitely the most important thing, right? Knowing what apps they live in, what daily tasks they constantly perform?” he said. “Companies realize that the more they have about you, the better the (AI) memory can be and the better they can personalize it.”

Claude Cowork is an app that “really shines” at this, says Witteveen; OpenClaw is another. The models are good enough to start making decisions for users and responding to instructions like: "You know a lot of things about me. You have all this context. Go and build the skills that will help me do a better job."

“With something like OpenClaw, you can customize it however you want, right? You can chat with it, you can say, ‘Hey, at 4 o’clock I want you to do this,'” Witteveen said.

However, token usage and security should always be kept in mind, he advised. OpenClaw has been plagued with security problemss since its launch. This has led many companies to uninstall the autonomous agent or ban its use entirely; However, these uninstalls must be done correctly so that IT leaders do not inadvertently delete their entire enterprise stack.

Meanwhile, in terms of token budget, customization can incur costs. “You need to think about the metrics you’re tracking,” Witteveen said. “This is very different from product to product, but the metrics around these things will be key."

Watch the podcast to learn more about:

• Why are companies not experimenting with AI skills right now? "it could be toast"

• How Zoom created an AI companion that tracks divergence of opinions, not just actions, in your meetings

• Why the question of build versus buy has become much more pressing for enterprise software

• Because "skills" may matter more than MCP for the future of enterprise AI

You can also listen and subscribe Beyond the pilot in Spotify, Apple or wherever you get your podcasts.

Source link