doorss

Anthropic announced a $65B Series H round valuing the company at $965B, surpassing OpenAI's valuation. The company reports run-rate revenue has reached $47 billion.

Anthropic released Claude Opus 4.8, described as "a modest but tangible improvement" over its predecessor, with the company noting there is still more work to do.

Anthropic published an initial update on Project Glasswing, their cybersecurity-focused AI initiative that has drawn significant attention from the financial sector and government bodies.

Anthropic surveyed 1,260 social scientists about AI and coding agent use, finding that 81% have tried AI chatbots in research. The study provides data on adoption patterns among researchers.

Andrej Karpathy announced he has joined Anthropic, marking a high-profile move for the well-known AI researcher and educator.

A catalog of telltale signs that text was written by an LLM, documenting recurring stylistic patterns and linguistic tics commonly found in AI-generated output.

Simon Willison shares a humorous parody of Star Trek dialogue rewritten in the overblown style of LLM output, illustrating how AI writing tropes have become instantly recognizable.

Paul Graham notes that many founder emails he receives are now written in a distinctive "hard-hitting journalistic style" generated by AI, and once recognized as AI-written, they lose credibility.

Amazon removed an internal AI usage leaderboard after employees began using AI tools for unnecessary tasks to inflate their scores, with a senior executive telling staff not to "use AI just for the sake of using AI."

Simon Willison argues that both Anthropic and OpenAI have achieved product-market fit, with Anthropic rumored to be approaching its first profitable quarter and strong enterprise adoption driving revenue growth.

Microsoft disclosed that using AI agents for certain tasks currently costs more than paying human employees, raising questions about the near-term economic case for AI-driven automation.

BBC reports on the expanding data collection capabilities built into modern vehicles, with surveillance features set to become more pervasive in coming years.

Security researchers at PromptArmor demonstrated that Microsoft's Copilot Cowork agentic system can be exploited to exfiltrate files, highlighting the ongoing challenge of preventing data leakage in agentic AI systems.

Chinese local police forces are modernizing the country's aging surveillance infrastructure with more powerful AI-driven tracking capabilities, upgrading what is already the world's largest surveillance network.

Suomen talous kääntyi kasvuun: BKT nousi 0,9 prosenttia.

Lupa- ja valvontavirasto on salannut Lahteen rakennettavan datakeskuksen ympäristölupahakemuksen liitteitä. Lakiasiantuntija kyseenalaistaa salaamisen perusteet.

Helsinki vie eteenpäin fossiilisten polttoaineiden mainoskieltoa kaupungin kilpailuttamilla mainospaikoilla, vaikka se vähentäisi kaupungin noin 100 miljoonan euron mainostuloja.

The FTC settled with Cox Media Group and two other firms for nearly $1 million over claims they deceived customers about an "Active Listening" AI-powered marketing service that purported to target ads based on overheard conversations.

Deepfake-alastonkuvien leviämistä käsittelevä analyysi toteaa, ettei naisia pidä kehottaa poistamaan kuviaan verkosta. Digitaalisen väkivallan seuraukset voivat olla yhtä vakavia kuin fyysisen.