DistantNews
Support us
🇲🇦 Morocco /Technology

European law requires AI companies to adopt new technology to detect AI-generated content

From Télégraphe · () Arabic

Translated from Arabic and summarized by DistantNews. Read the original for the full story.

At a glance

Explainer Sources not specified New plan
  • AI companies are expected to use hidden mathematical and statistical patterns in generated text so specialized tools can identify AI-produced content.
  • Reports cited by the article say major systems, including Claude, ChatGPT and Gemini, have begun incorporating such markers.
  • The requirements are linked to the EU AI Act, with full compliance required by December 2, 2026, according to the report.

Artificial intelligence companies are moving toward embedding invisible watermarks in text generated by their systems. The markers are designed to help distinguish machine-produced writing from human work.

The watermark does not appear as visible letters or symbols. Instead, it takes the form of a hidden mathematical and statistical pattern connected to how a model selects words. To readers, the text should still look natural, while specialized detection tools could identify it as AI-generated.

The article says the move comes as requirements under the European Union’s AI Act begin to take effect. Reports cited in the article say different AI models must achieve full compliance by December 2, 2026.

Major systems, including Claude, ChatGPT and Gemini, have reportedly begun adding the markers to the texts they produce. The technology is expected to affect people who directly copy and paste AI-generated material, including for homework, academic research and message preparation.

About this summary

Originally published by Télégraphe in Arabic. Translated, summarized, and contextualized automatically by DistantNews, with a note on how the source frames the story. Not individually reviewed before publishing. How this works.

More Stories

From Our Blog