Proposed benchmark: AI passes if it can fix Microsoft Azure and Google Ads interfaces

A proposal has surfaced suggesting that frontier LLMs (large language models—AI systems trained on vast text data) should be tested by deploying them to improve notoriously complex interfaces: Microsoft's Azure UX and Google Adwords UX. Success would mean substantial reduction in the 'dreadfulness' users experience when navigating these platforms.

The test works by letting an AI agent (software that decides and executes actions on its own) take control of each interface and optimize it autonomously. If Microsoft and Google's latest AI models can meaningfully simplify these historically difficult user experiences, it would validate industry claims that today's frontier models are genuinely capable.

For business professionals, students, and anyone using Azure or Ads, this matters because it reframes AI capability claims: instead of abstract benchmarks, the real test is whether AI can solve concrete, frustrating problems in tools millions already use daily. A passing grade would mean these interfaces could become materially easier to navigate.