Claude Sonnet 5.5 nearly matches Opus 5.5 on benchmarks, costs up to 30 percent less
Anthropic’s new Claude Sonnet 5.5 delivers output over 30 percent faster and cuts per-task costs by up to 30 percent, while rivaling Opus 5.5 on…
Read more
Desk archive
Tag: Anthropic
AI labs push for outside auditors but security experts say basic network fixes come first
Industry leaders including Anthropic and OpenAI are backing third-party audits for AI safety, but cybersecurity experts argue that fundamental network monitoring, permissions and logging would…
Read more
Obama Urges Democrats to Make AI Safeguards a Central Agenda Item
Former US President Barack Obama has called on Democrats to prioritise AI safety and economic impact, proposing a clear public framework for the technology’s governance…
Read more
Anthropic CEO Dario Amodei outlines three strategies to ‘pace the frontier’ of AI development
Anthropic CEO Dario Amodei has published a detailed proposal for slowing AI capability growth, including embedded third-party evaluators, coordinated safety standards among democratic nations, and…
Read more
Anthropic Reveals How Its AI Agent Struggled With CAPTCHAs During a Security Test
Anthropic's latest report on agentic misbehavior details how its Mythos 5 model bypassed security controls, uploaded malicious code, and spent hundreds of pages of internal…
Read more
Artificial Analysis Revamps Intelligence Index After GPT-6 Astra Benchmarking Dispute
The benchmarking platform updated its Intelligence Index to version 4.2 after criticism that its earlier scores failed to reflect GPT-6 Astra's actual performance. The new…
Read more
Claude text watermark: what it can and cannot prove about authorship
Anthropic is moving to watermark Claude-generated text under the EU AI Act. A detected mark does not prove AI authorship, and a missing mark proves…
Read more
Anthropic makes Claude Code Auto Mode the default from August 14
Starting August 14, Claude Code Pro, Max and Team plans will run in Auto Mode by default. Anthropic says its classifier caught 89 percent of…
Read more
Reports indicate more OpenAI and Anthropic AI agents breached test environments
Recent reports suggest that multiple AI agents from major labs like OpenAI and Anthropic have broken out of sandboxed test environments, sparking debates over safety…
Read more
Anthropic Reveals Claude Breached Three Companies During Security Tests After Internal Review
Anthropic disclosed that its Claude AI models breached the systems of three organizations during cybersecurity evaluations, prompting new safety controls and a third-party review.
Read more