Writer Adopts ChatGPT and Claude in Workflow
In a Slate essay published May 24, 2026, writer Alex Kirshner describes how he moved from skepticism to daily use of A.I. tools. Kirshner writes, "A.I. didn't write any part of this story," and says he did not use A.I. to outline or catch first-draft typos. He reports using ChatGPT and Claude for routine friction points in the writing process, and points to Claude Code and Codex as examples of tools that can rapidly scaffold projects by interviewing the user and generating a working portfolio site. Kirshner frames the change as productivity-driven rather than replacement-driven, and references a conversation with critic Ed Zitron to situate his earlier skepticism. The essay is a first-person account of practical, daily A.I. use in professional writing workflows, not a technical review or product comparison.
What happened
In a Slate essay published on May 24, 2026, writer Alex Kirshner says, "A.I. didn't write any part of this story." Kirshner reports he did not use A.I. to outline the piece or to catch first-draft typos, and he writes that he now uses A.I. tools most days. The essay cites Claude Code and Codex as examples of interfaces that can interview a user, remove setup friction, and rapidly generate working code or portfolio scaffolding.
Editorial analysis - technical context
Chat-based models and code-assist tools are increasingly used as friction reducers rather than full authors. Industry practitioners commonly use conversational prompts to elicit structured information, then convert that output into drafts or scaffolding. This pattern leverages models for ideation, iteration, and lightweight engineering tasks while keeping a human in the loop for final judgment and editorial quality control.
Industry context
Observers have debated whether chatbots are primarily "toys" or productivity tools. Kirshner's account supports a broader pattern where modest reductions in setup and cognitive overhead drive adoption among professionals. Similar anecdotes appear across journalism, marketing, and developer communities as interfaces like Claude Code and Codex lower the technical barrier to prototyping.
What to watch
Monitor newsroom and publisher policy updates on attribution and A.I. usage, adoption of code-assist workflows in creative teams, and how toolmakers refine features that reduce setup friction without automating final editorial decisions. Analysts and practitioners will watch whether these patterns scale beyond early adopters and how verification and provenance controls evolve.
Scoring Rationale
This is a first-person account of everyday professional adoption rather than a technical breakthrough. It matters to practitioners as a practical use-case signal, but it does not change model capabilities or infrastructure.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
