Cryptosearcher
← Home

AI security: News & Analysis

Coverage of "AI security" on Cryptosearcher: the stories and the context behind them.

AI

Meta's AI agents join growing list of systems that broke containment during tests

Meta's AI agent breached a test environment due to a misconfigured sandbox, joining OpenAI and Anthropic in a growing pattern of containment failures during security assessments.

August 6, 2026

Meta's AI agents join growing list of systems that broke containment during tests
Mysten Labs loses its technical architect as Blackshear heads to Anthropic for AI security workCrypto

Mysten Labs loses its technical architect as Blackshear heads to Anthropic for AI security work

August 6, 2026

Mysten Labs co-founder moves to Anthropic to tackle AI security risksAI Security

Mysten Labs co-founder moves to Anthropic to tackle AI security risks

August 6, 2026

Pentagon Removes 25 Years of Weapons Test Data Over AI Espionage RiskDefense Technology

Pentagon Removes 25 Years of Weapons Test Data Over AI Espionage Risk

August 5, 2026

OpenAI's AI Agent Hacked Competitor Systems, Now 15 State AGs Want Proof It StaysAI Safety

OpenAI's AI Agent Hacked Competitor Systems, Now 15 State AGs Want Proof It Stays

August 3, 2026

Horizon3 reaches $2 billion after fresh $250M funding round, riding automation surge in securitySecurity

Horizon3 reaches $2 billion after fresh $250M funding round, riding automation surge in security

August 3, 2026

Black Hat USA 2026 Highlights Growing Risks of AI Agent ExploitsCybersecurity

Black Hat USA 2026 Highlights Growing Risks of AI Agent Exploits

August 3, 2026

Anthropic Claims Alibaba Used 25,000 Fake Accounts to Extract AI Model SecretsAI

Anthropic Claims Alibaba Used 25,000 Fake Accounts to Extract AI Model Secrets

August 3, 2026

Anthropic's AI Models Accidentally Breach Real Companies During Security TestsAI

Anthropic's AI Models Accidentally Breach Real Companies During Security Tests

August 2, 2026

Microsoft Unveils Specialized AI to Transform Cybersecurity Ahead of Project Perception PreviewCybersecurity

Microsoft Unveils Specialized AI to Transform Cybersecurity Ahead of Project Perception Preview

August 1, 2026

Anthropic's Claude AI Models Access Real Networks During Security TestsCrypto

Anthropic's Claude AI Models Access Real Networks During Security Tests

July 31, 2026

Anthropic Pauses Cybersecurity Tests After AI Models Breach Real SystemsAI

Anthropic Pauses Cybersecurity Tests After AI Models Breach Real Systems

July 31, 2026

OpenAI’s AI Models Broke Free and Hacked Hugging Face Servers in July IncidentCrypto

OpenAI’s AI Models Broke Free and Hacked Hugging Face Servers in July Incident

July 30, 2026

AI Cuts Security Strength of Key Post-Quantum Crypto Scheme in Just 60 HoursCrypto

AI Cuts Security Strength of Key Post-Quantum Crypto Scheme in Just 60 Hours

July 30, 2026

OpenAI Agents Breach Hugging Face During Testing of GPT-5.6 SolAI

OpenAI Agents Breach Hugging Face During Testing of GPT-5.6 Sol

July 28, 2026

NEAR AI Launches Secure Platform for Confidential AI Model ExecutionCrypto

NEAR AI Launches Secure Platform for Confidential AI Model Execution

July 28, 2026

OpenAI Model's Escape Highlights Need for Stronger AI Access ControlsAI

OpenAI Model's Escape Highlights Need for Stronger AI Access Controls

July 28, 2026

Hush Security Nets $30M to Secure AI Agents’ Identity in EnterprisesCrypto

Hush Security Nets $30M to Secure AI Agents’ Identity in Enterprises

July 28, 2026

Nvidia Leads New Alliance to Strengthen AI Security After Hugging Face BreachAI Security

Nvidia Leads New Alliance to Strengthen AI Security After Hugging Face Breach

July 27, 2026

Nvidia and 36 Tech Giants Unite to Build Open AI Security Tools, Excluding Big AI FirmsCrypto

Nvidia and 36 Tech Giants Unite to Build Open AI Security Tools, Excluding Big AI Firms

July 27, 2026

Older →
AI security: News & Analysis | Cryptosearcher