Working Notes on Agent Systems/Brad Zhang

@teach_fireworks / X longform

Microsoft AI Survey: The more powerful AI is, the more precious human judgment is.🔥 AI continues to expand

Microsoft AI Survey: The more powerful AI is, the more precious human judgment is.🔥 AI continues to expand the boundaries of human capabilities, but it also increases...

May 5, 2026 · 2 min read

Microsoft AI Survey: The more powerful AI is, the more precious human judgment is.🔥 AI continues to expand the boundaries of human capabilities, but it also increases the value of high-quality judgment.

Microsoft's AI survey hit the core truth!

Most AI users know this well: as AI undertakes more and more work, the two most important human core skills in our hearts are at the top of the list - 50% believe that the quality of AI output is controlled, and 46% choose critical thinking, that is, the ability to analyze information objectively and make rational judgments.

More noteworthy is that 86% of respondents said that they would only treat AI output as a first draft, not a final answer, and always insist on being responsible for thinking about the results.

The human role has long since shifted from “generating answers” to “evaluating, optimizing, and pocketing answers.” And the frontline workforce is even more aware of this!

In terms of all relevant indicators such as critical thinking and AI quality control, their recognition far exceeds that of ordinary workers, and their work methods are also more sober: ✅ 43% of frontier practitioners deliberately do some work without AI, only to keep their abilities online.

The proportion of ordinary workers is only 30%.

✅ 53% of people deliberately stop and think before work, dividing the boundaries between AI and people.

Only 33% of ordinary workers never give the right to think to AI.

They know that long-term success Key: Always polish the core human capabilities and never let them degrade and shrink in the AI era.

# AIAge # HumanCoreCompetitiveness # WorkplaceInsights

Visual summary

Article argument map

Generated from the post's content graph

FORMATTOPICCAPABILITYMARKETcoverscoverssignalssignalsFORMATlongform noteTOPICai workbenchesTOPICevaluationCAPABILITYAI-native workbenchCAPABILITYevaluation
Mermaid outline
flowchart LR
  format-long_post["longform note"]
  topic-ai-workbenches["ai workbenches"]
  topic-evaluation["evaluation"]
  capability-ai-native-workbench["AI-native workbench"]
  capability-evaluation["evaluation"]
  format-long_post -->|covers| topic-ai-workbenches
  format-long_post -->|covers| topic-evaluation
  format-long_post -->|signals| capability-ai-native-workbench
  format-long_post -->|signals| capability-evaluation

Visual structure

Essay structure map

Built from summary and key paragraph positions

Microsoft AI Survey: The more powerful AI is, the more precious human judgment is.🔥...THESISMicrosoft AI Survey:The more powerful AIis, the more precioushuman judgment is.🔥SIGNALMicrosoft AI Survey:The more powerful AIis, the more precioushuman judgment is.🔥OPERATORIn terms of allrelevant indicatorssuch as criticalthinking and AIIMPLICATION# AIAge #HumanCoreCompetitiveness# WorkplaceInsights
Mermaid outline
flowchart LR
  thesis["Microsoft AI Survey: The more powerful AI is, the more precious human judgment is.🔥 AI continues to expand..."]
  signal["Microsoft AI Survey: The more powerful AI is, the more precious human judgment is.🔥 AI continues to expand..."]
  operator["In terms of all relevant indicators such as critical thinking and AI quality control, their recognition far..."]
  implication["# AIAge # HumanCoreCompetitiveness # WorkplaceInsights"]
  thesis -->|frames| signal
  signal -->|develops| operator
  operator -->|lands in| implication

Source: View the original post