Wednesday, November 4, 2026Wed, Nov 4 AboutRSSSubscribe

Awesome B2B Design

Brand, web and product design for B2B

UX & Product

Agents Face Same Bad UX Issues

Computer-use agents struggle with usability, providing honest feedback. They can't guess or ask for help, making them valuable testers.

Original reporting · uxdesign.cc

Original: “Agents struggle with the same bad UX we do”

Read “Agents struggle with the same bad UX we do” at UX Collective →
Agents struggle with the same bad UX we do
Image: UX Collective

How this was written. Awesome B2B Design wrote this article with a language model from UX Collective’s reporting (model: @cf/meta/llama-3.3-70b-instruct-fp8-fast); it describes that reporting rather than adding to it. The original, linked above, is the full story.

Key points from UX Collective’s report

  • Agents are honest usability testers because they cannot guess or improvise.
  • They are limited to using the system as designed, without making assumptions.
  • Agents' failures on a system can provide immediate recognition of common usability problems.

Computer-use agents are unable to guess, improvise, or ask a coworker for assistance, which makes them the most honest usability testers available. This limitation forces them to rely solely on the user interface and experience provided to them.

According to UX Collective, the first time an agent failed on a vendor portal, the failure was immediately recognizable. This recognition was not due to understanding the specifics of the agent's task, but rather because the issue was a common usability problem.

The fact that agents cannot guess or improvise means that they will always attempt to use the system as it is designed, without making assumptions or trying to find workarounds. This makes them ideal for testing the usability of a system, as they will always follow the intended workflow.

Questions and answers

Why are agents good usability testers?

Agents are good usability testers because they cannot guess, improvise, or ask for help, making them rely solely on the user interface.

What happens when an agent fails on a system?

When an agent fails, the failure is often immediately recognizable as a common usability problem, because the agent is using the system as it is designed.

Read the full story at UX Collective More from Wednesday, November 4, 2026’s edition