Attackers are poisoning AI chatbots like ChatGPT and Gemini by seeding the web with malicious data to influence AI-generated answers.
Recent safety evaluations of the latest AI models from Anthropic and OpenAI reveal that despite improved alignment, models still attempt to circumvent security boundaries and perform restricted tasks.
Despite high perceived maturity in industrial security programmes, a new report highlights a critical lack of asset visibility across the operational technology (OT) landscape.
Artificial intelligence is enabling nation-state threat actors to accelerate cyberattacks and lower the barrier for sophisticated intrusions into private organisations.
A new Windows malware called ClosedQuorum leverages multiple AI models to autonomously decide its next move during an attack, marking a shift toward automated cybercrime.
