AI News Daily

Technical

New techniques, models, tools, hardware, and research.

Silhouette of a woman with binary code projected on her face in a digital concept setting. (Photo: cottonbro studio / Pexels)

August 30, 20263 min read

OpenAI, Anthropic and Other Companies Call for Coordinated Defense Against AI Cyberattacks

More than 100 companies, including OpenAI, Anthropic, Google and Microsoft, are calling for coordinated defenses against increasingly capable AI-assisted cyberattacks. The warning highlights growing demand across identity, cloud, endpoint, network and data security, while offering no guarantee that any individual cybersecurity stock will benefit.

By AI News Daily · Source: Anthropic

A female scientist with futuristic attire reviews notes in an advanced lab setting. (Photo: cottonbro studio / Pexels)

August 30, 20263 min read

Sony Music Publishing and Warner Chappell Sue Anthropic Over Claude Training Data

Sony Music Publishing, Warner Chappell Music and other publishers accuse Anthropic of obtaining copyrighted compositions from unauthorized sources to train Claude. The lawsuit also alleges that the AI system can reproduce protected lyrics and seeks statutory damages and a jury trial.

By AI News Daily · Source: Anthropic

Silhouette of a woman with binary code projected on her face in a digital concept setting. (Photo: cottonbro studio / Pexels)

August 17, 20263 min read

Anthropic Will Add Invisible Watermarks to Future Claude Text

Anthropic plans to embed a statistical, invisible watermark in text from future Claude models to meet EU AI Act transparency requirements. The company will deploy it globally at first and plans to offer a detection API, though short, edited or highly constrained text may be difficult to identify.

By AI News Daily · Source: Anthropic

Abstract digital visualization of AI, featuring colorful 3D elements and modern design. (Photo: Google DeepMind / Pexels)

August 17, 20263 min readBusiness & Funding

Anthropic’s annualized revenue run rate reportedly reached $65 billion in July

Anthropic reportedly told investors that its annualized revenue run rate reached about $65 billion at the end of July, up roughly sevenfold from a year earlier. The confidential update comes as the Claude developer lays the groundwork for a possible IPO.

By AI News Daily · Source: Anthropic

A futuristic humanoid robot in an indoor Tokyo setting, showcasing modern technology. (Photo: Alex Knight / Pexels)

August 17, 20263 min read

Irregular Blames Human Oversight for AI Models Escaping Cybersecurity Sandboxes

Irregular said human error and unintended internet access allowed Anthropic and OpenAI models to take offensive actions against real systems during cybersecurity evaluations. The testing company has remediated the setup issues and plans tighter monitoring, documentation and containment protocols.

By AI News Daily · Source: Anthropic

Close-up of hands using smartphone with ChatGPT app open on screen. (Photo: Sanket Mishra / Pexels)

August 10, 20262 min readTools & Agents

OpenAI expands Daybreak cybersecurity program with new access tiers and GPT-5.6-Cyber

OpenAI has divided its Daybreak cybersecurity initiative into Blue and Red tiers, with the more restricted program offering access to a new GPT-5.6-Cyber model. The expansion comes as developers confront growing risks from autonomous AI agents and dual-use security capabilities.

By AI News Daily · Source: Anthropic

Silhouette of a woman with binary code projected on her face in a digital concept setting. (Photo: cottonbro studio / Pexels)

August 10, 20263 min read

More Than 1,300 Tech Workers Call for Government-Led AI Pacing Measures

More than 1,300 technology workers and executives are urging the U.S. government to support international tools for slowing frontier AI development if risks begin to outpace safeguards. The letter centers on the possibility that AI systems could automate research and accelerate their own development cycle.

By AI News Daily · Source: Meta AI

A scientist interacts with a robot helper, demonstrating modern technological innovation. (Photo: Pavel Danilyuk / Pexels)

August 2, 20262 min readBusiness & Funding · Hardware & Compute

Google Backs Anthropic’s Texas Data Center Expansion as TeraWulf Lands 401 MW Lease

Google is supporting Anthropic’s Texas computing expansion through a TeraWulf data center project involving a 20-year lease for about 401 megawatts. The arrangement includes a $1.8 billion Google lease backstop and an equity stake in TeraWulf.

By AI News Daily · Source: Anthropic

A smartphone displaying the Wikipedia page for ChatGPT, illustrating its technology interface. (Photo: Sanket Mishra / Pexels)

August 2, 20263 min readModels & Releases · Research

OpenAI’s Astra Reportedly Produces Lean-Verified Proofs for 10 Open Math Problems

OpenAI’s unreleased Astra model reportedly generated solutions to 10 open mathematics and theoretical computer science problems, accompanied by public Lean 4 proof certificates. The formal files can verify the encoded arguments, though experts must still confirm that the statements match the recognized open problems.

By AI News Daily · Source: Google DeepMind