OpenAI Watermarking Text, AI Risks, and Global Tech Advances: OpenAI, Anthropic, and China's AI Surge

Podcast: AI in Marketing · Stan Berteloot

Opublikowano: 2026-10-06

OpenAI plans to introduce an invisible watermark for AI-generated text in the EU, complying with the EU AI Act's transparency rules. This watermark subtly alters word choices to create a detectable pattern, remaining invisible to readers. A technical report details the method, textGrain, but notes that editing can reduce its effectiveness. Access to the detector will initially be limited to approved researchers. Executives from AI companies testified at a New York City Council hearing on AI risks and potential legislation. Concerns were raised about AI-related harm and the need for regulation beyond self-regulation. The hearing highlighted the...