Reuters ·

Anthropic says it found 5,500 verified vulnerabilities in April-October and Glasswing partners found 129K+ in April-July; 33K+ were critical or high severity

Anthropic is expanding a program that allows vetted cybersecurity professionals to test its most powerful AI models with fewer safeguards …

In this story Anthropic Glasswing

Anthropic says it found 5,500 verified vulnerabilities in April-October and Glasswing partners found 129K+ in April-July; 33K+ were critical or high severity

Lead Source

Reuters — Anthropic says it found 5,500 verified vulnerabilities in April-October and Glasswing partners found 129K+ in April-July; 33K+ were critical or high severity ↗

How this story grew

Coverage · 0 Discussion · 0

Oct 6Oct 8

More

Fast Company: Fast Company

CSO: CSO

Help Net Security: Help Net Security

The Decoder: The Decoder

SecurityWeek: SecurityWeek

Security Affairs: Security Affairs

Bloomberg: Bloomberg

Discussion

@choblin29 @andrewcurran_ @adnanthekhan @iam_zachi @lainshawty @pandare__ @azakasekai_ @jun_song @ustas_eth @cjav_dev @andrewmohawk @anthropicai

Related stories

Axios: OpenAI, Anthropic, and researchers are probing tens of thousands of frontier model security incidents, including sandbox escapes and website hijacking

Ars Technica: Microsoft's September 2026 Patch Tuesday fixes a record ~972 vulnerabilities, taking its total flaws patched in 2026 so far to 2,760, more than double from 2025

Krebs on Security: Microsoft releases updates fixing 1,400+ security vulnerabilities, surpassing September's record of 972, which was already ~10x the normal monthly patch load

Anthropic: Claude flagged 493 security issues across 25 open-source projects, covering 5M+ lines of C, C++, Java, PHP, and Python code; one went unnoticed for 21 years

BleepingComputer: Citrix confirms two critical NetScaler zero-day RCE vulnerabilities are being exploited in attacks, says it has released security updates to fix the flaws

Discuss with AI

ChatGPT ↗ Claude ↗

← Back to TechSnif