PulseAugur
EN
LIVE 14:31:36

OpenAI rolls out text watermarking to comply with EU AI Act

By PulseAugur Editorial · [3 sources] ·

OpenAI is implementing text watermarking for its models, including ChatGPT and Codex, to comply with EU regulatory requirements under the EU AI Act. While the technology aims to help identify AI-generated content, OpenAI acknowledges its significant limitations, noting that watermarks can be undetectable, especially in short passages, and can be removed through rewriting or translation. The company is making text watermarking available to API customers worldwide and will initially deploy it for eligible text in the EU. AI

IMPACT This move by OpenAI signals a growing trend of AI companies integrating content provenance features to meet regulatory demands and address concerns about AI-generated content.

RANK_REASON Product feature launch with regulatory compliance angle.

Read on X — OpenAI →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

OpenAI rolls out text watermarking to comply with EU AI Act

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Product feature launch with regulatory compliance angle.
Source corroboration
3 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
policy, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
Standard
On-topic for AI-industry coverage; kept in the public index.
Story freshness
1 days old
Coverage has settled into its steady-state source set.

Full methodology in our editorial standards.

COVERAGE [3]

  1. X — OpenAI TIER_1 English(EN) · OpenAI ·

    Watermarks have limits. They’re often undetectable, especially in short passages. Rewriting or translating text can completely remove the watermark.

    Watermarks have limits. They’re often undetectable, especially in short passages. Rewriting or translating text can completely remove the watermark. That’s why, for now, only approved researchers will have access to our detector, so they can help us evaluate and improve the

  2. X — OpenAI TIER_1 English(EN) · OpenAI ·

    When possible, we want to give people a choice—and be clear about what watermarking can and can’t do.

    When possible, we want to give people a choice—and be clear about what watermarking can and can’t do. A text watermark helps answer one question: “Was this likely generated by an OpenAI model?” It works by embedding an invisible statistical signal in text as it’s generated. A

  3. X — OpenAI TIER_1 English(EN) · OpenAI ·

    We're expanding our approach to content provenance to include text in response to EU regulatory requirements, while recognizing the significant limitations of c

    We're expanding our approach to content provenance to include text in response to EU regulatory requirements, while recognizing the significant limitations of current text watermarking technology. Our tools already help verify whether an image or audio file was created with our

RELATED ENTITIES

RELATED TOPICS