Imagine a future where apps are mainly used by AI agents, not humans. By 2032, we might spend more time interacting with these agents than using traditional apps. This shift is being driven by advancements in AI-powered autonomy, as highlighted by a recent study.

One key area is agentic systems, which can write and use code accurately. Companies are quickly incorporating these capabilities to speed up engineering tasks. But agents aren't ready to handle all aspects of a digital core yet. They can only tackle bits of it.

The future also holds promise for generative UI, where AI can create personalized user interfaces. This is a big change from the past, where one UI had to fit all users.

To start, companies should experiment with internal agents. Begin with specific tasks and gradually expand their functions and data access. This will prepare companies for external-facing agents.

However, as these agents become more common, companies must ensure they're trustworthy. They need to monitor the agents and have clear guardrails in place. Transparency is key to building employee trust in these systems.

While AI agents are impressive, they're not perfect. They're expensive to run, can't always provide clear explanations, and aren't fully predictable. But with the right grounding, like code and functions, they can become more explainable and trustworthy.

AI Agents: The Next Big Thing in App Use

Actions