Preds logoPreds

Stop guessing. Read the whole assignment.

We trained Preds on student essays too. Paste a Hungarian paper to see the overall AI signal and the sentences worth a closer look.

For Hungarian essays and assignments
Overall result and sentence-level view
Try your own text or choose an example
Examples:
0 characters

Why Preds?

Trained on student essays. Built for teachers.

Student essays are part of the training data. When a paper raises questions, Preds shows the flagged sentences in the complete submission.

Made for Hungarian

The model was trained with Hungarian essays and other educational writing. You can check a paper in the language it was written.

Sentences, not just scores

Review signals inside the text as well as the overall result. If only a few paragraphs raise questions, you can examine them without labeling the entire assignment.

A more useful conversation

Ask about a source, draft, or line of reasoning where a passage was flagged. The student’s explanation and writing process still matter.

Preds result with sentence-level highlights in Hungarian text

In practice

The full text stays in view.

You do not need to match a separate list of warnings back to the assignment. Preds shows the overall estimate beside the text and highlights sentences that may deserve attention. You can return to the student’s original argument while reviewing the result.

  1. 01Paste an assignment, or start a check from your account.
  2. 02Read flagged sentences in the complete paper.
  3. 03Compare the result with sources, earlier work, and the student’s explanation.

A detector gives an estimate, not proof of authorship. Do not base a grade or disciplinary decision on the result alone.

Guides and research.

Fresh articles about research, operations, and advice in the world of AI detection.

All articles
01
EmberiAI szövegOct 2, 2026

Nem, az LLM-ek nem utánozzák az embert!

Csábító azt gondolni, hogy mivel az LLM-eket az emberi szöveg modellezésére tanítják, a kimenetük megkülönböztethetetlenné válik az emberi írástól. A valóságban azonban az LLM-ek nem csupán megközelítik az emberi szöveget: utótanítással (post-training) olyan segítőkész asszisztensekké alakítják őket, amelyek követik az utasításokat, ez pedig egy szűk tartományra korlátozza az írásukat, amely felismerhető marad.

Read article