Anthropic - it-daily.net - Daily relevant IT news for decision-makers

New Anthropic Policy

Claude Users Face Mandatory ID Verification

17. June, 2026 - 15:15

The AI company Anthropic is introducing identity verification for users of its chatbot Claude starting in July. The move comes amid increasing regulatory pressure and tightening export control requirements.

Between outrage and political pressure

After Claude Fable 5 Ban: Europe Reacts to Sudden AI Model Shutdown

17. June, 2026 - 07:14

On June 12 at 5:21 p.m. Eastern Time, Anthropic received a US government export control directive. Within hours, Claude Fable 5 and Mythos 5 were disabled worldwide. The move triggered a wave of political and industry reactions across Europe.

Unclear Threat Assessment

Anthropic Blocks Claude Fable 5 Access Following Government Security Order

16. June, 2026 - 07:10

Something unprecedented has happened: OpenAI rival Anthropic has been forced to suddenly restrict access to its most advanced AI model after the U.S. government raised national security concerns. Many details remain unclear.

AI safety controls become transparent

Anthropic Reveals Hidden AI Limits After Developer Outcry

15. June, 2026 - 07:50

Anthropic has revised its approach to communicating safety restrictions in its new language model Claude Fable 5, following mounting criticism from developers and AI researchers. The move comes after concerns that the company had been applying hidden model downgrades when users triggered certain types of requests.

Access to Offensive AI Model

Anthropic Opens Mythos AI Hacking Model to EU Review

03. June, 2026 - 05:50

After weeks of exclusion, Anthropic has offered the European Commission and cybersecurity agency ENISA access to its controversial security-focused AI model, Mythos.

Processing incomplete data sets in AI systems

Anthropic Releases Claude Opus 4.8 with Improved Data Validation

31. May, 2026 - 05:58

The new AI model Claude Opus 4.8 reduces false claims and orchestrates complex IT migrations within systems through Dynamic Workflows.

Automated Vulnerability Chaining

Too Dangerous to Release? Cloudflare Warns Over Anthropic’s Cyber AI “Mythos”

22. May, 2026 - 07:07

After conducting internal tests, Cloudflare is warning that Anthropic’s new AI model Mythos may be capable of autonomously chaining together harmless software flaws into severe cyberattacks.

Power users

New usage limits for Claude subscribers

29. July, 2025 - 07:18

Anthropic has announced new weekly usage limits for its AI assistant Claude in order to curb the intensive use of the coding tool Claude Code. This mainly affects users who run the system around the clock, so-called power users.

Partly legal, partly illegal

Judge: Anthropic may train AI with books without author’s permission

25. June, 2025 - 09:11

In a landmark ruling, a US federal court in San Francisco has decided that the use of copyrighted books to train an AI model is legal under certain conditions.