Skip to content
Säkerhet· UpdateAvailable

Anthropic introduces invisible watermarks to all Claude output

Anthropic is introducing invisible watermarks and signed metadata to all content generated by Claude. The move is an adaptation to the EU AI Act and is being rolled out globally.

By the Aheadline editorial team·11 aug. 2026·2 min read·Source: Entity-watch: AnthropicVerifierad signalAI-generated
Anthropic introduces invisible watermarks to all Claude output
Anthropic introduces invisible watermarks to all Claude output
By · Policy- & EU-reporter
Last updated

What happened?

AI company Anthropic is introducing invisible watermarks and signed metadata for all content created by the Claude AI model. This measure ensures that machine-generated or processed text will contain machine-readable traces, enabling the identification of its origin. The technology encompasses Claude via the website, the API platform, Claude Code, Claude Cowork, Claude Tag, and cloud-based services.

Key facts

EU AI Act-deadline2 augusti 2026
Tekniska metoderOsynlig vattenstämpel och signerad metadata
TillämpningsområdeGlobalt (Webb, API, Claude Code, molntjänster)

Why it matters

The initiative follows Anthropic's commitment to the EU AI Act's code of conduct regarding transparency for AI-generated content (Article 50.2). The goal is to increase transparency surrounding AI-created material, facilitate the detection of automatically generated text, and counteract the spread of unclear or misleading content on the internet.

Who is affected?

The decision impacts all developers, businesses, and private individuals using Claude's services globally. Content creators and systems developers can expect text generated via Anthropic to carry invisible markers that can be identified by external tools.

Impact on the EU

The labelling is being introduced as a direct response to requirements in the EU AI Act (Article 50.2). Models launched in the EU from 2 August 2026 will support the machine-readable labelling from their first day of release, but the functionality is being rolled out globally.

What else you should know

The system combines two distinct methods for traceability. The first is an invisible watermark embedded directly into the AI text, and the second is signed metadata that verifies the content was created or processed by Claude.

Frequently asked questions

Quick answers about this story

Vad har hänt?
Anthropic tillkännagav att de inför osynliga vattenstämplar och signerad metadata på allt innehåll som skapas av deras Claude-modeller.
När hände det?
Beskedet lämnades i samband med att Anthropic undertecknade EU AI Acts koder för transparens, med krav på stöd i EU-modeller från den 2 augusti 2026.
Varför spelar det roll?
Åtgärden gör det möjligt att identifiera AI-genererad text globalt, vilket ökar transparensen och hjälper användare och system att spåra innehåll som skapats av Claude.
Påverkar det EU?
Ja, kraven baseras på EU AI Act men Anthropic väljer att rulla ut märkningen globalt över alla sina plattformar och API-tjänster.
Original source
Entity-watch: Anthropic·cybersecuritynews.com

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

Verifierad signal

Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.

AI-verktyg i artikeln

Topics

#Ethics#EU AI Act#Safety#Anthropic#Policy
[ 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 introduces invisible watermarks to all Claude outp"