In Amanda Caswell’s test of one old comedy draft, Claude was the best overall writing coach for her goal: it caught a story-structure gap the other assistants missed and explained the issue without taking over the fix. ChatGPT stood out for line-level rhythm and recurring imagery, while Gemini offered the clearest, easiest-to-apply suggestions. That is one writer’s judgment about one draft—not a universal ranking.
What happened when all three assistants got the same draft?
Tom’s Guide AI Editor Amanda Caswell found an early comedy piece in Google Drive and gave the same draft and instructions to ChatGPT, Claude and Gemini. She wanted to know what was working, where the story dragged and how to improve the comedy without losing her voice. She was not looking for the most aggressive rewrite or sentence polishing for its own sake.
As an Amazon Associate I earn from qualifying purchases.
The story’s workplace technology included instant messaging, Clippy and a fax machine. Caswell wanted the assistants to respect that period setting rather than modernize it. She also asked them to preserve her voice and limit their feedback to three recommendations—a constraint she felt encouraged prioritization and restraint. Her account, published August 27, 2026, describes a qualitative comparison, not a controlled benchmark or a scored test. Read Caswell’s account.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Why was Claude her best overall writing coach?
Claude noticed that readers never learned what the employees in the story actually did for work. Caswell saw that as a structural gap: the missing context affected the story’s logic. She valued that Claude explained the problem and left her to decide how to address it, rather than supplying the joke or rewriting the passage for her. It also recognized why a seemingly small detail, such as a character’s toothbrush, mattered to the story.
That combination—looking beyond individual sentences while leaving creative decisions with the writer—made Claude her overall pick. It does not establish that Claude will be the strongest structural editor for other genres, drafts or writers.
What did ChatGPT and Gemini do better?
ChatGPT: rhythm, pacing and imagery
Caswell found ChatGPT strongest as a line editor. It paid attention to rhythm, pacing and recurring imagery, including an inventive suggestion to return to a Tetris image. Like Gemini, it suggested tightening the opening, trimming description and expanding dialogue. If a draft already makes sense structurally but needs attention to how the prose moves, those were the strengths that stood out in her example.
Rank #2
- Multilingual Handwriting Recognition Write naturally with the wireless writing pad instead of typing. Accurately recognizes handwritten Traditional Chinese, Simplified Chinese, English, Japanese, numbers, symbols, and mixed-language input for seamless text entry.
- Write Smarter with AI Boost your productivity with the built-in AI Writing Assistant. Draft emails, rewrite content, summarize documents, translate text, and generate ideas faster with the help of AI.
- Personalized Digital Signature Sign PDF documents, forms, contracts, and emails with your own handwritten signature, giving your digital documents a more professional and personal touch.
- Handwriting input to MS Word, MS PowerPoint, Google Docs, WeChat, Whatsapp, Line, and more. Win/Mac supported
- Plug & Play Wireless Convenience Simply connect the included wireless USB receiver and start using immediately—no driver installation required. Compatible with Windows and macOS for effortless setup.
Gemini: a clear revision path
Gemini’s recommendations were the easiest for Caswell to absorb and apply. It identified strong moments and offered concrete suggestions, such as tightening the opening and making dialogue punchier. It did not catch the story-logic gap Claude identified in this draft.
Which assistant should you choose for your own draft?
Use Caswell’s result as a guide to the kind of feedback to request, not as a permanent leaderboard. The assistants helped in different ways in her example:
- For a possible structural problem: Claude was the one that identified the missing work context in Caswell’s story and explained why it mattered.
- For sentence rhythm and recurring images: ChatGPT’s feedback drew her attention to pacing and a potential Tetris callback.
- For straightforward, actionable edits: Gemini’s suggestions were the clearest and easiest for her to apply.
- For protecting your voice: State that you want critique rather than a rewrite, and ask the assistant to preserve the setting, voice and details that carry meaning.
These are observations about one comedy draft. The comparison does not include repeat trials, a scoring rubric or statistical results, so it cannot show how consistently each assistant performs across different writers or kinds of work.
How can you get feedback without having the draft rewritten?
Give all three assistants the same material and instructions if you want to compare their advice. Ask for a limited number of prioritized recommendations, explain what must remain intact, and request an explanation of each suggested change. For example, you could ask for three high-impact notes on structure, pacing and clarity; specify that the assistant must preserve your voice and period details; and ask it to identify any detail it thinks should stay rather than cutting it automatically. This approach reflects the constraints Caswell said helped keep feedback focused.
Rank #4
Read each recommendation against your purpose for the piece. An AI may suggest cutting a detail because it seems unnecessary, even when that detail is part of what makes the story distinctive. Treat feedback as advice to assess, not an instruction to accept.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Where can you work with the draft?
Editing workflows are separate from the quality of the advice in Caswell’s comparison. Official product documentation describes ways to work with drafts, but those features do not establish which assistant gives better writing feedback.
ChatGPT
OpenAI says writing blocks can hold editable drafts, let users request changes to selected text or the whole draft, and support undoing or redoing recent AI-assisted edits. Availability and behavior can depend on plan, device, workspace settings, model and feature rollout. Check OpenAI’s Canvas documentation.
Claude
Anthropic documents Artifacts as a space for self-contained content such as documents, which users can edit by chatting or directly in the artifact. Its help page lists Artifacts for Free, Pro, Max, Team and Enterprise plans; templates and related features can have additional plan or settings requirements. Check Anthropic’s Artifacts documentation.
Gemini in Docs
Google’s March 2026 announcement described Gemini in Docs features for drafting with selected relevant files, refining highlighted sections or a whole document, and matching a writing style or reference-document format. Google described those features as beta and said they were rolling out to Google AI Ultra and Pro subscribers at the time. Check the product for current availability. Read Google’s announcement.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteQuick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
