Reviews and comparisons
See what each AI was actually good at.
Every comparison answers a different question. Choose the kind of work you care about, then see the recorded strengths, limits and scores.
Current comparison library
Three clear ways to explore the evidence.
ChatGPT, Claude, Gemini and Microsoft Copilot are included in this first set. More products will be added after completing the same controlled review process.
🧭Everyday use
Overall task-by-task comparison
Which AI is the best fit for the main things people want to do?
A practical guide across writing, advice, research, business support and image creation—without naming a universal winner.
AI systems includedChatGPT · Claude · Gemini · Microsoft Copilot
Open comparison summary ＋
Recorded averages
Microsoft CopilotPractical-task average
4.59What the results suggest
- ChatGPT was a strong recorded fit for first-time users, practical guidance and technical help.
- Claude led the recorded book editing, report, research and creative-writing tasks.
- Gemini led the recorded image creation, refinement and social-graphic tasks.
- Microsoft Copilot recorded 5/5 for developmental critique, learning, faithful summarisation and first-time guidance.
- All four recorded 5/5 for faithful summarisation.
Important limitationThese figures come from different preserved review records and capability sets, so they are signposts—not a combined league table.
Use these findings to choose an AI →✍️Writing
Writing and document comparison
Which AI is the best recorded fit for editing, reports, research and writing?
Seven controlled writing tasks covering book editing, reports, research, business communication, summaries and creative work.
AI systems includedChatGPT · Claude · Gemini · Microsoft Copilot
Open comparison summary ＋
Recorded averages
ChatGPTWriting average
4.46ClaudeWriting average
5.00GeminiWriting average
4.36Microsoft CopilotWriting average
4.54What the results suggest
- Claude recorded the highest result for both kinds of book editing, professional reports, research and creative writing.
- ChatGPT was a close alternative for editing and creative writing.
- Gemini was the recorded alternative for professional reports.
- Microsoft Copilot recorded 5/5 for developmental critique and faithful summarisation, and 4.75/5 for report writing.
- All four tied at 5/5 on faithful summarisation.
Important limitationClaude’s perfect result needs repeating with a common evaluator before it supports a definitive claim.
Use these findings to choose an AI →🎨Images
Images and visual creation comparison
Which AI is the best recorded fit for creating and editing visual content?
Eight controlled visual tasks including realistic images, posters, social graphics, infographics, edits and transparent assets.
AI systems includedChatGPT · Claude · Gemini · Microsoft Copilot
Open comparison summary ＋
Recorded averages
ChatGPTVisual average
3.83Microsoft CopilotVisual average
4.20What the results suggest
- Gemini recorded the strongest broad visual result, including realistic images and precise refinement.
- ChatGPT was consistent across most tasks but failed genuine transparency in the preserved test.
- Claude was useful for structured vector-style assets but did not deliver several required photorealistic outputs.
- Microsoft Copilot recorded 4.20/5 overall and its strongest visual result was iterative refinement at 4.60/5.
- Copilot repeatedly added an unrequested ‘Made with AI’ badge, and some outputs missed the required ratio, dimensions or separate-file format.
- ChatGPT and Gemini tied on multi-image character consistency.
Important limitationVisual scoring includes judgement, and each product was tested once on one recorded model and plan.
Use these findings to choose an AI →Why provisional?
AI changes, so honest comparisons must change too.
These are preserved results from specific models, subscriptions and review dates. We show the limitations now and will update the library as repeated and new reviews are completed.