Skip to content
Anthropic· UpdateGlobal

Anthropic adds invisible watermarking to Claude models to meet EU requirements

Anthropic is expanding its invisible text watermarking to older Claude models, such as Opus and Sonnet. The global rollout is intended to meet EU requirements regarding the traceability of AI-generated text.

By the Aheadline editorial team·3 nov. 2026·2 min read·Source: Entity-watch: AnthropicAggregerad källaAI-generated
Anthropic adds invisible watermarking to Claude models to meet EU requirements
Anthropic adds invisible watermarking to Claude models to meet EU requirements
Anthropic adds invisible watermarking to Claude models to meet EU requirements
By · Policy- & EU-reporter
Vad betyder det för mig?

What happened?

Anthropic is implementing invisible text watermarking across its broader range of AI systems, meaning text created with Claude will contain a pattern that can be identified automatically. The feature was introduced in August on the company's newest models and is now being extended to older systems, including Mythos 5, Opus 4.7, Opus 4.6, and Sonnet 4.6. The watermarking is being activated globally for all users with no option to opt out.

Key facts

Omfattade äldre modellerMythos 5, Opus 4.7, Opus 4.6, Sonnet 4.6
Opt-out-möjlighetIngen (obligatorisk globalt)
Regulatorisk drivkraftEU-reglering för generativ AI

Why it matters

The measure is being implemented to comply with EU regulations requiring the identification and labelling of AI-generated content. By including older model versions, Anthropic ensures that its entire fleet of models meets the legal requirements for transparency.

Who is affected?

All global users and developers generating text with Claude are directly affected by the change. Organisations and educational institutions requiring verification of whether text content was generated by Anthropic models will now have access to a tool for automatic detection.

Impact on the EU

The technology was launched in direct response to EU requirements for generative AI traceability and applies automatically to all users within the union. Because the watermarking cannot be disabled, Anthropic is meeting the requirements across all global markets.

What else you should know

The rollout is occurring in phases across Anthropic's product portfolio, following an initial focus solely on the developer's newest AI models.

Frequently asked questions

Quick answers about this story

Vad har hänt?
Anthropic rullar ut osynlig textvattenmärkning bredare över sina Claude-modeller, inklusive äldre system som Mythos 5 och Opus- samt Sonnet-versioner.
När hände det?
Utrullningen till äldre modeller meddelades den 2 november 2026, efter att tekniken först introducerats i augusti 2026.
Varför spelar det roll?
Förändringen genomförs för att möta EU:s regulatoriska krav på att generativ AI ska kunna märkas och spåras automatiskt.
Påverkar det EU?
Ja, vattenmärkningen omfattar samtliga användare globalt och lanserades specifikt för att följa EU:s lagstiftning.
Original source
Entity-watch: Anthropic·uk.finance.yahoo.com

The link opens in a new window and leads to the publisher's own site.

Aggregerad källa

Källan är en aggregator eller syndikering — vi rekommenderar att verifiera hos primärutgivaren.

AI-verktyg i artikeln

Topics

#Anthropic#AI Act#Models
[ STAY UP TO DATE ]

Get similar news straight to your inbox

No affiliate linksCancel anytimeGDPR-friendly
[ Frequency ]
[ What do you want to read about? ]

You'll receive updates on 2 topics.

The reader's room

Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.

Sign in to submit a comment or question.

Loading comments…
How this affects you

Read the article through your role

  • Decide whether this affects strategy over 6–12 months or is just noise.
  • Discuss with leadership: do we own the right question or does ownership need to move?
  • Ask: what risk are we taking by NOT acting on this this quarter?

Generated angle — not editorial analysis of "Anthropic adds invisible watermarking to Claude models to me"