Google DeepMind Proposes AI ‘Monitors’ to Police Hyperintelligent Models
DeepMind’s approach to AGI safety and security splits threats into four categories. One solution could be a “monitor” AI.
Artificial Intelligence
Serial Entrepreneurs Raise $43M to Counter AI Deepfakes, Social Engineering
Adaptive is pitching a security platform designed to replicate real-world attack scenarios through AI-generated deepfake simulations.
The post Serial Entrepreneurs Raise $43M to Counter AI Deepfakes, Social Engineering appeared first on SecurityWeek.
Vulnerabilities Expose Jan AI Systems to Remote Manipulation
Vulnerabilities in open source ChatGPT alternative Jan AI expose systems to remote, unauthenticated manipulation.
The post Vulnerabilities Expose Jan AI Systems to Remote Manipulation appeared first on SecurityWeek.
Cyberhaven Banks $100 Million in Series D, Valuation Hits $1 Billion
Cyberhaven bags $100 million in funding at a billion-dollar valuation, a sign that investors remain bullish on data security startups.
The post Cyberhaven Banks $100 Million in Series D, Valuation Hits $1 Billion appeared first on SecurityWeek.
Apple Rolls Out iOS 18.4 With New Languages, Emojis & Apple Intelligence in the EU
Apple’s iOS 18.4 lets EU users choose default navigation apps like Google Maps or Waze, complying with the Digital Markets Act for more competition and user control.
AI Giving Rise of the ‘Zero-Knowledge’ Threat Actor
The rise of zero-knowledge threat actors powered by AI marks a turning point in the business of cybercrime where sophisticated attacks are no longer confined to skilled attackers.
The post AI Giving Rise of the ‘Zero-Knowledge’ Threat Actor appeared fi…
Google DeepMind Unveils Framework to Exploit AI’s Cyber Weaknesses
DeepMind found that current AI frameworks are ad hoc, not systematic, and fail to provide defenders with useful insights.
The post Google DeepMind Unveils Framework to Exploit AI’s Cyber Weaknesses appeared first on SecurityWeek.
Microsoft Using AI to Uncover Critical Bootloader Vulnerabilities
Using the Security Copilot tool, Microsoft discovered 20 critical vulnerabilities in widely deployed open-source bootloaders.
The post Microsoft Using AI to Uncover Critical Bootloader Vulnerabilities appeared first on SecurityWeek.
Developers Wanted: OpenAI Seeks Feedback About Open Model That Will Be Revealed ‘In the Coming Months’
Find out how to provide OpenAI with your input about its upcoming open language model, which Sam Altman stated will be a “reasoning” model like OpenAI o1.
PortSwigger Launches Burp AI to Enhance Penetration Testing with AI
PortSwigger, the makers of Burp Suite, has taken a giant leap forward in the field of cybersecurity with the launch of Burp AI, a groundbreaking set of artificial intelligence (AI) features designed to streamline and enhance penetration testing workflo…