Claude Articles

46 articles
Filter by importance: AI Rating: High AI Rating: Medium AI Rating: Low
Filter by specialty: Chat & General Models Image Generation Video Generation Voice & Music Coding Search & Agents
Claude News 09/28/2026 Read AI Rating: High

Anthropic Launches Claude Sonnet 5.5, 30% Faster and Up to 30% Cheaper Than Sonnet 5

Anthropic introduced Claude Sonnet 5.5, jumping from 10.3% to 70.6% on Terminal-Bench 4.0 and scoring just two points below Opus 5.5 on the GDPval-AA real-world occupational benchmark, while running over 30% faster and up to 30% cheaper per task at identical per-token pricing.
Claude News 09/25/2026 Read AI Rating: Medium

Anthropic Opens a Directory Submission Portal for Claude Plugins

Anthropic launched a self-serve portal for developers to submit MCP connectors and plugin bundles to the Claude directory, complete with auto-validation, review tracking, and post-launch install analytics.
Claude News 09/23/2026 Read AI Rating: High

Anthropic's New Life Sciences Lab: Claude Agents Discover a Novel Enzyme System

Claude agents searching a DNA sequence database for 21 hours identified array-associated reverse transcriptases (ART), a previously unrecognized enzyme system in bacteriophages resembling CRISPR arrays, in early results from Anthropic's new life sciences research lab.
Claude News 09/22/2026 Read AI Rating: High

Anthropic Launches Claude Opus 5.5, Matching Fable 5.1 at 40% Lower Cost

Anthropic's first model in the new 5.5 family posts big agentic-coding gains, cuts API pricing 20%, and ships with an 85% drop in containment-boundary circumvention on internal audits.
Claude News 09/18/2026 Read AI Rating: High

Anthropic and Accenture Commit $1 Billion Over Five Years to Embedded AI Safety Evaluation

Anthropic partnered with Accenture's Faculty division to embed independent evaluators inside Anthropic with employee-level access to observe model training and deployment decisions, backed by a five-year, $1 billion-plus joint investment, with more non-exclusive evaluator partnerships planned alongside existing work with METR.
Claude News 09/17/2026 Read AI Rating: Medium

Anthropic Launches a Verification Program Giving Life Sciences Researchers Fewer-Interruption Access to Claude

Anthropic's Life Sciences Verification Program lets vetted biology researchers use Claude's Mythos, Opus, and Sonnet models with offline misuse monitoring instead of real-time blocking, offering standard (annual, team-wide) and high-risk (six-month, project-specific) access tiers.
Claude News 09/10/2026 Read AI Rating: High

Anthropic's September Threat Report: AI Has 'Collapsed the Labor Gap' for Cyberattackers

Anthropic's latest threat intelligence report documents eight months of disrupted misuse across cyber operations, influence campaigns, and AI-supply-chain attacks, warning that multi-agent frameworks now let Claude execute entire attack chains with minimal human oversight.
Claude News 09/01/2026 Read AI Rating: Medium

Anthropic Launches Enterprise Frontier Safeguards, Built With Goldman Sachs, Stripe, and 100+ Others

Anthropic introduced Enterprise Frontier Safeguards, letting customers store Claude interaction logs in their own cloud accounts under their own encryption keys while keeping fully automated, cross-session misuse detection, developed with over 100 customers including all nine ARC member banks.
Claude News 09/01/2026 Read AI Rating: High

Anthropic Launches Claude Fable 5.1 and Mythos 5.1, Cutting Cache Read Costs 75%

Anthropic released Claude Fable 5.1, generally available with major gains in agentic coding and scientific research, alongside Mythos 5.1 โ€” a technically identical model with more permissive safeguards restricted to verified cybersecurity and life-sciences researchers.