Sobering up on "AI"
An HCI retrospective on LLMs, mental models, the operator-seat transition, and what happens when human-factors training meets vibe coding.
Parts3
Span2024–2026
TrackAI & Automation
- 01
Sobering up on "AI": A Human-Computer Interaction Retrospective (Part 1)
Why the AI wave echoes the Three Mile Island moment for human factors, and why LLM interfaces recreate the command-line problem under conversational veneer.
- 02
Sobering up on "AI": Why designing with LLM is Confusing (Part 2)
The inversion of mental models. LLMs spoof cognitive affordances on the inanimate side, doubling HCI's workload instead of reducing it.
- 03
Sobering up on "AI": A Year of Designing With LLMs (Part 3)
AI was going to double the design field's work, or so I claimed. A year later I have a different report. The doubling claim survived. The future it assumed did not.