Apple introduced Apple Reference Image, a new standard for iPhone 18 Pro models that uses signed sensor data to create an uneditable reference picture alongside an editable version, aiming to help ver…
Apple introduced a new Reference camera mode for the iPhone 18 Pro at its fall event, which digitally authenticates that an image hasn't been altered with AI by creating an unalterable reference photo…
Apple announced Wednesday during its Surprise and Shine event that it is introducing Apple Reference Image, a feature for the iPhone 18 Pro that uses signed sensor data and Apple's Private Cloud Compu…
Apple announced at WWDC 2026 that it will embed Google DeepMind's SynthID invisible watermarks into AI-generated and AI-edited images across its ecosystem, starting with iOS 27 later this year. By mid…
Anthropic announced that its Claude models will soon embed invisible watermarks in text output, while Google made its visible Gemini watermark on generated images optional, highlighting divergent appr…
Anthropic and other LLM providers have begun adding an imperceptible watermark to AI-generated text, sparking debate over whether it degrades output quality. An explainer breaks down the watermarking …
An open-source utility with 5.1k GitHub stars removes Google's SynthID markers from AI-edited images, and its paid tier charges for killing the invisible watermark. The exploit is a pip install, and t…
Google released the Omni 1.1 Flash video model, adding 1080p and 4K output options to the Gemini Omni Flash model, which previously generated video at 720p. The multimodal model, unveiled at Google I/…
Anthropic introduced a watermark into Claude's output this month, and others have followed or will soon follow suit. The watermark, similar to Google's SynthID, uses statistical bias from a pseudorand…
A developer built a minimal, educational version of Google's SynthID-Text style watermarking system, demonstrating how token probabilities are manipulated to embed a detectable signal in LLM output. T…
Anthropic has begun watermarking outputs of its Claude AI model using a scheme based on Google's SynthID and the Gumbel Softmax method proposed by Scott Aaronson in 2022. Aaronson, who credits Anthrop…
AI companies are buying pallets of old printed books to avoid training on AI-generated 'slop,' according to ISBNdb, a broker sourcing bulk print acquisitions for AI labs. Meanwhile, Google announced o…
A developer presented Google's Gemini Omni, a natively multimodal AI model for video generation and editing, at a GDG Calabar session. The model supports conversational editing, audio synthesis, and i…
Anthropic has introduced text watermarking in its Claude LLM outputs, a move that has sparked backlash from AI enthusiasts, according to David Gerard's coverage. The watermarking, which encodes a sing…
Within four hours of Anthropic confirming that Claude models would globally embed invisible, machine-readable watermarks into AI-generated content, developer Guillaume Meyer published code to remove t…
Anthropic announced that text created with Claude, its large language model, will now contain an invisible watermark, making it detectable by A.I. detectors at the individual word level. The change, m…
Major foundation model vendors, including Anthropic and Google, deployed statistical text watermarking and cryptographic metadata standards to comply with the EU AI Act Article 50, which took effect A…
John Gruber, writing on Daring Fireball, argues that Anthropic's watermarking of AI-generated text in Claude inherently degrades prose quality, despite claims to the contrary, and objects to its secre…
Google is letting users of its Gemini and Flow apps turn off the visible watermark on AI-generated images, video, and music, while keeping invisible SynthID watermarks and C2PA metadata embedded. The …
Anthropic plans to release text watermarking based on Google DeepMind's SynthID, which deterministically modulates the pseudorandom number generator behind token selection to allow later detection of …