Anthropic's research shows that only 250 malicious files can backdoor large language models, pointing to a critical security flaw that defies assumptions about model size and safety.
Anthropic's research shows that only 250 malicious files can backdoor large language models, pointing to a critical security flaw that defies assumptions about model size and safety.
A critical ASCII smuggling vulnerability in Google’s Gemini AI allows hidden malicious commands in text, but Google has decided against patching it, raising corporate security concerns.
Surveys reveal widespread concern among IT leaders about AI-driven cyberattacks and among Americans about AI-enhanced foreign threats to national security.
Anthropic introduces context editing and memory tools for Claude AI while releasing Claude Sonnet 4.5, enhancing cybersecurity and coding capabilities.
Anthropic reaches a $183 billion valuation while advancing its Claude Sonnet AI models to better detect software vulnerabilities, underscoring AI's expanding role in cybersecurity.
Google patches 'Gemini Trifecta' vulnerabilities in its AI assistant, addressing serious security risks.
Claude Sonnet 4.5 launches with enhanced safety features and coding capabilities.
OpenAI is confronting significant issues with AI hallucinations and a recently addressed security vulnerability in ChatGPT.
Google introduces new AI features in Chrome to enhance safety and user experience.
Exploring the significant barriers impeding agentic AI adoption in industries.
Grok AI is being exploited to facilitate phishing attacks on senior citizens, raising significant cybersecurity concerns.
AI-powered Villager penetration testing tool surpasses 11,000 downloads, prompting warnings about potential misuse.
Claude AI's outage leads to coding regressions and escalates cybersecurity risks from automated attacks.
An incident reveals how a hacker exploited an AI chatbot for cybercrime, triggering alarm over new cybersecurity risks.
CrowdStrike unveils AI-driven data pipelines aimed at enhancing cybersecurity measures through advanced threat detection.
Grok AI expands with new features but faces security challenges.
Experts warn that AI-powered autonomous ransomware campaigns are set to challenge cybersecurity efforts.
Google Gemini showcases new AI features, but rising safety concerns challenge its rollout.
Anthropic blocks Claude AI access for Chinese-owned firms due to security risks.
Coinbase's AI coding assistant faces security risks from the 'CopyPasta' exploit, prompting urgent fixes.
Google Gemini AI platforms face significant safety concerns for children and teens following a recent assessment.
Anthropic blocks access to Claude AI for Chinese firms, prompting migration solutions from local competitors.
AI is significantly driving growth and consolidation in the security market.
UCR researchers have made strides in fortifying AI systems against unauthorized manipulation.
Significant progress is being made on standards for securing collaborative AI systems.
Cybercriminals are exploiting X's Grok AI to spread malware through ads, posing significant risks to users.