The tool below is the free end of that and uses no model at all.
WeSolve+ reads the whole document and writes the questions for you
Upload your PDF, photograph your notebook, or point the camera. WeSolve+ writes questions from that material, explains why each answer is right, reads the chapter back to you as a podcast, and remembers every item you missed until you own it.
The tool below is a small browser-only tool and it is not WeSolve+: paste a few lines and text rules turn them into cards on the spot. The real app, the one that uses AI, is behind the link above.
Notes to practice
This is a browser-only tool, and that is all it isIt splits the text you paste by rule, and nothing else. WeSolve+ is a different thing entirely: it reads your whole PDF with AI, writes the reasoning behind every question, speaks the chapter back to you, and remembers what you missed so it can return it. Try the real app now, free!
This tool uses text rules, not AI. There is no model behind it, which is deliberate on a page about AI: the useful thing here is a rule you can read in one paragraph, and knowing which parts of your work genuinely need more than that. Your text stays in the browser and is never sent anywhere.
Three uses, and only one of them is contested
Producing work you submit is the contested one, and every institutional policy is written about it. Understanding something is not contested at all: asking for a concept explained three ways is closer to a textbook than to cheating. Practising is the third and it is where the most value sits, because turning your own material into questions produces something to be tested on rather than something to hand in. Sorting a request into one of these three before you type it settles almost every case, and it is a cleaner test than any list of banned tools. The definitions institutions actually enforce are set out under academic integrity, and the specific offence they name is plagiarism, presenting somebody else's work as your own.
Disclosure is usually the whole policy
Most institutions have moved from prohibition to declaration, because prohibition was unenforceable and everybody knew it. What they now ask is that you say what you used and what for, and the sentence they want is boring: which tool, which part of the work, and what you did with the output. Where a course bans it outright the ban is generally on assessed work rather than on studying. Detection tools exist and are unreliable in both directions, producing both false accusations and easy evasion, which is precisely why the sensible policies stopped relying on them. Read your own course's wording before you paste anything, because the variation between departments is larger than the variation between institutions.
How to check an output in under a minute
Verification is a habit rather than an attitude, and three checks catch most of what goes wrong. Run the number: substitute the result back into the condition it came from, because a long chain of correct method can still contain one quiet slip. Open the source: a reference with an author, a year and a title can still describe a paper that was never written, which is the failure named hallucination. And match the route to your own course, since an approach your department does not teach loses marks even when it arrives somewhere true. None of this is scepticism about generative systems; it is the same thing you would do with a friend's notes, and skipping it is how automation bias gets you.
The skill that is actually being tested
A student who can tell a good answer from a fluent one is in a different position from a student who cannot, and that difference will not be visible until an exam. What separates them is ordinary critical thinking applied to a new surface, which is the substance of digital literacy rather than anything novel. Practically it comes down to two questions asked of every output: what would make this wrong, and how would I find out. Both are answerable in under a minute and both are the thing your degree is supposed to teach you anyway.
The use with the best return, and it is unglamorous
Turning material you already have into questions you have to answer is the highest value use, and it is the one nobody argues about. It is also supported by the strongest evidence in the study literature: the review by Dunlosky and colleagues rated practice testing and distributed practice high while summarising and highlighting were rated low, and the Roediger and Karpicke experiments found the advantage of being tested grew as the delay to the final test increased. An assistant that hands you an answer bypasses exactly the mechanism those studies measured.
What the app does, and what it will not do
Upload a PDF, a photograph of a page or a camera shot and WeSolve+ writes an explained quiz, a card deck and a short audio recap from that file. It works on material you already hold rather than producing new work to submit, which is the distinction this page has argued for throughout. Questions you miss are kept aside for the next session. Free to begin on the web and on iPhone and iPad, metered by file rather than by question, with no separate Android build. Figures on pricing.
Sources used on this page
- Academic integrity
- Plagiarism
- International Center for Academic Integrity
- Generative artificial intelligence
- Critical thinking
- Digital literacy
- Hallucination in artificial intelligence
- Educational technology
- Formative assessment
- Active recall
- Testing effect
- Spaced repetition
- Desirable difficulty
- Roediger and Karpicke, Test-Enhanced Learning (2006)
- Karpicke and Roediger, the critical importance of retrieval (2008)
- Cepeda et al., distributed practice meta analysis (2006)
- Dunlosky et al., improving students' learning (2013)
- What Works Clearinghouse, organizing instruction and study
- The Learning Scientists, retrieval practice
- Retrieval Practice, the research library
- Worked-example effect
- Cloze test
- Anki manual, importing text files
- MDN, the textarea element
- MDN, the details element
- MDN, the Clipboard API
- W3C, Web Content Accessibility Guidelines 2.2
- WeSolve+ on the App Store
- Ethics of artificial intelligence
- Automation bias
- Information literacy
- Prompt engineering
| What you ask for | Which use | What happens |
|---|---|---|
| Write my essay on this | Producing | Misconduct, and no learning either |
| Explain this concept three ways | Understanding | Uncontested, and often better than the book |
| Argue against my draft | Understanding | Uncontested, and the best use of a model |
| Give me the answer to question four | Producing | Marks lost twice: now and in the exam |
| Turn my chapter into questions | Practising | The use with the most evidence behind it |
| Check my working for a slip | Understanding | Fine, and worth verifying yourself |
Is using AI for schoolwork cheating?
It depends entirely on which of three things you are doing: producing work you submit, understanding something, or practising. Only the first is contested, and every policy is written about it.
Do I have to declare that I used it?
Usually yes, and that is now most of what policies ask. Which tool, which part of the work, and what you did with the output. Read your own course's wording, since departments vary more than institutions.
Can detectors tell?
Unreliably, in both directions. They produce both false accusations and easy evasion, which is exactly why the sensible policies stopped depending on them and asked for disclosure instead.
What is the single best use for a student?
Turning material you already have into questions you have to answer. It is uncontested, and it is the use with the strongest evidence behind it in the study literature.
Does the tool on this page use AI?
No, and that is deliberate on a page about AI. It is a text rule you can read in one paragraph, which makes it a useful reference point for what genuinely needs a model.
Is anything I paste uploaded?
No. There is no server behind this box. Watch the network panel while you type, or disconnect entirely and it still works.
Last updated: 2026-08-15