Updated
Updated · O'Reilly Media · Jun 24
AI Industry Shifts to Agent Experience as 97.1% of 856 MCP Tools Show Quality Issues
Updated
Updated · O'Reilly Media · Jun 24

AI Industry Shifts to Agent Experience as 97.1% of 856 MCP Tools Show Quality Issues

3 articles · Updated · O'Reilly Media · Jun 24

Summary

  • Agent Experience is emerging as the new focus for AI teams, reframing agent strategy around how systems are discovered, understood and used rather than around MCP, Skills or other fast-changing protocols.
  • A 2026 Queen’s University study of 856 tools across 103 MCP servers found 97.1% had at least one quality issue and 56% failed to state their purpose clearly, suggesting design—not protocol—was the main weakness.
  • The AX approach centers on five practices: identify which agents customers use, map delegated use cases, test where agents fail, iterate product behavior, and automate validation with scoring tools such as AXIS in CI.
  • The shift matters as agent adoption spreads: BCG and MIT Sloan found 35% of organizations already use agentic AI and another 44% plan to, raising pressure to make agent interactions a durable competitive advantage.

Insights

With AI agents becoming the new user, is the entire discipline of human-centered design now obsolete?
Beyond fleeting protocols, who will build the platforms that govern our future AI-driven economy?
Is the AI 'tool trap' creating a silent crisis of developer burnout and wasted innovation?

From Protocols to Agent Experience: 79% of Enterprises Embrace Agentic AI—Trends, Risks, and the 2028 Outlook

Overview

In 2026, artificial intelligence reached a turning point as Agent Experience (AX) emerged as a vital design focus, moving beyond protocol-centric development. This shift was driven by a growing recognition that the success of AI agents depends on the quality of their interactions and usability, not just technical protocols. Highlighted in John Maeda’s 2026 Design in Tech Report, researchers and enterprises began prioritizing agent interaction quality as a core engineering concern. As agentic AI usage surged and organizations deployed autonomous agents across diverse functions, the industry realized that delivering great experiences for both users and agents is now essential for staying competitive.

...