What is the same
All three are hosted chat UIs from OpenAI, Anthropic, and Google. All can hallucinate and invent citations. Keep passwords, unpublished research, and other people’s private data out of them.
Free tiers and study modes change. Check each product’s current limits before you pay.
How people usually split them
ChatGPT: everyday brainstorming, quick explanations, and the default many tutorials assume.
Claude.ai: long readings and careful instruction-following.
Gemini: Google Search grounding and Docs/Drive workflows.
Three prompts, three apps
A: explain a concept you know.
B: outline a seven-day plan with real constraints.
C: critique your own outline and list missing evidence.
Keep the better answers in your notes.
Policy
Some workplaces and courses ban generators. Some require disclosure. Some allow outlining only.
AI detectors are imperfect. Your written policy decides what is allowed.
FAQ
Which one is free?
All three offer free tiers with limits. Paid plans raise caps and unlock features.
Should I subscribe to all three?
No. Pick one primary for two weeks. Use free tiers of the others only for a short side-by-side test.
Where do NotebookLM and Perplexity fit?
NotebookLM is for chatting against sources you provide. Perplexity is for cited web research. See those comparisons when your bottleneck is sources, not general chat.