SpaceXSI quietly flips Grok API text watermarking to off by default
The thorny issue of watermarking AI-generated text took an interesting turn this past week, with SpaceXSI seemingly U-turning on plans The post SpaceXSI quietly flips Grok API text watermarking to off by default appeared first on The New Stack .
SpaceXSI, formerly known as SpaceXAI, has reversed its decision to automatically watermark text generated by its Grok AI models. The company's documentation, which initially stated that watermarking would be enabled by default starting from November 7, 2026, now indicates that the feature is off by default. Developers must manually enable the watermarking feature through their team's administration settings on the Grok API.
The change occurred between November 9 and November 10, 2026. This decision comes amid increasing transparency obligations for AI system providers under the EU AI Act, which will take full effect on August 2, 2026. SpaceXSI joins other major AI labs like Anthropic, OpenAI, and Google in implementing text watermarking, though each company has taken a distinct approach to the feature.
Written by urgent.news from The New Stack's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
