A general chat assistant plus one note-taking tool covers about 90% of real value. Most of the hundreds of specialised tools are thin wrappers around the same models, and their features get absorbed by the main assistants within months.
What genuinely helps, by use case:
**Understanding difficult material** — this is the strongest use. Paste a passage you don't understand and ask for it at a simpler level, then ask follow-ups until it clicks. Ask for analogies. Ask 'what's the intuition behind this?' Having an infinitely patient explainer that never makes you feel stupid for asking a third time is a genuine improvement over textbooks and, honestly, some lecturers.
**Testing yourself** — ask it to generate practice questions from your notes, then answer *before* looking. Ask it to grade your explanation and identify what you missed. This works because it's active recall, which is the highest-value study technique, and the main barrier to active recall has always been the effort of making questions.
**The Feynman technique, with a partner** — explain a concept to it and ask it to identify gaps or errors in your explanation. This is the single best study use and almost nobody does it.
**Drafting and editing** — first drafts, restructuring an argument, tightening prose, generating outlines. Much better as an editor of your thinking than a replacement for it.
**Summarising material you provide** — meeting notes, long documents, research papers. Reliable because it's grounded in text you supply rather than recalled from training.
**Coding** — explanation, debugging, boilerplate, tests.
What doesn't help, despite the marketing:
- Anything that reads material *for* you so you can skip it. The reading was the learning; a summary of a chapter you never read produces the feeling of knowledge without any.
- Tools that generate flashcards you never review. The generation was never the bottleneck.
- Most 'AI-powered' study apps that are a chat box with a subject-specific prompt.
- Note-taking tools you spend more time configuring than using — a real and common trap.
The discipline that determines whether any of this helps: use it to increase the effort you put into thinking, not to reduce it. Asking it to explain something and then testing yourself is more effortful than reading a textbook, and that's why it works. Asking it to write your assignment is less effortful, and you'll learn nothing — the feeling of productivity and the fact of learning come apart very easily here.