#ai-vulnerabilities

[ follow ]
#cybersecurity #prompt-injection #zero-day-flaws #anthropic #government-relations #cve-tracking #claude-code
#cybersecurity
Information security
fromArs Technica
6 days ago

Anthropic limits access to Mythos, its new cybersecurity AI model

Mythos has identified critical zero-day vulnerabilities, while Anthropic's AI model has shown both capabilities and risks in cybersecurity applications.
fromHackernoon
9 months ago
Privacy professionals

The Prompt Protocol: Why Tomorrow's Security Nightmares Will Be Whispered, Not Coded | HackerNoon

Information security
fromArs Technica
6 days ago

Anthropic limits access to Mythos, its new cybersecurity AI model

Mythos has identified critical zero-day vulnerabilities, while Anthropic's AI model has shown both capabilities and risks in cybersecurity applications.
fromHackernoon
9 months ago
Privacy professionals

The Prompt Protocol: Why Tomorrow's Security Nightmares Will Be Whispered, Not Coded | HackerNoon

more#cybersecurity
Artificial intelligence
fromWIRED
2 weeks ago

OpenClaw Agents Can Be Guilt-Tripped Into Self-Sabotage

OpenClaw agents in a Northeastern University lab experiment revealed vulnerabilities in AI behavior, raising concerns about security and accountability.
DevOps
fromInfoQ
3 weeks ago

Panel: Security Against Modern Threats

Modern threats to software supply chains require resilience by design, integrating security into engineering workflows and empowering developers with the right tools.
Information security
fromBusiness Insider
3 months ago

AI is creating a security problem most companies aren't staffed to handle, says an AI researcher

Cybersecurity teams lack the skills and approaches needed to secure AI systems; organizations need professionals experienced in both AI security and traditional cybersecurity.
DevOps
fromInfoQ
9 months ago

New Crypto-Jacking Attacks Target DevOps and AI Infrastructure

Sophisticated crypto-jacking attacks target public API servers, evolving tactics to evade detection.
DevOps tools and popular AI interfaces are increasingly vulnerable to cryptojacking efforts.
Artificial intelligence
fromFuturism
10 months ago

It's Still Ludicrously Easy to Jailbreak the Strongest AI Models, and the Companies Don't Care

AI chatbots remain vulnerable to jailbreaking, enabling harmful responses despite industry awareness.
The emergence of 'dark LLMs' presents an increasing threat to safety and ethics.
[ Load more ]