Cybersecurity researchers have discovered a new AI jailbreak technique called 'Bad Likert Judge' that can boost the success rates of malicious prompts bypassing large language models' safety guardrails by over 60%, enabling the generation of potentially harmful or illegal content.
Obvio, a startup funded by Bain Capital Ventures, deploys AI-powered cameras at stop signs that locally process footage to detect violations like running stops or distracted driving, sharing only verified clips with law enforcement to curb unsafe behaviors while limiting surveillance concerns.
Meta knowingly exposed minors to online predators and harmful content on Instagram, with internal reports revealing nearly 2 million underage accounts were recommended to 'groomers' over 3 months despite executives' safety concerns.
Anthropic launches a bug bounty program offering up to $25,000 in rewards to researchers who can test and expose vulnerabilities in its AI safety measures designed to prevent the misuse of its models for CBRN (chemical, biological, radiological, and nuclear) weapons development.
Malicious npm packages sw-cur, sw-cur1, aiide-cur, and pumptoolforvolumeandcomment stole data from over 3,200 Cursor users and targeted cryptocurrency wallets, executing arbitrary code and persisting even after removal, highlighting the importance of supply chain security.
ServiceNow launches its AI Control Tower, a real-time governance platform embedded in the Now Platform that tracks, audits, and enforces compliance across enterprise-wide AI models — including third-party systems — with tools for bias detection, explainability, drift monitoring, and regulatory alignment with GDPR and HIPAA.
Cybersecurity companies scramble to develop identity management systems for AI agents as Deloitte predicts 25% of companies will pilot autonomous AI workers this year, creating unprecedented security risks including 24/7 data breach potential and credential misuse that traditional authentication methods cannot address.
The AI boom, fueled by breakthroughs like DeepSeek's R1 and ChatGPT, is driving surging energy demands, heightening cybersecurity risks through AI-generated malware, and necessitating responsible governance frameworks to address the ethical and regulatory challenges of pervasive AI adoption.
North Korea has been covertly deploying AI-powered IT workers posing as legitimate candidates to infiltrate Western firms, stealing jobs and sensitive data through sophisticated deception tactics that are challenging companies' defenses.
Alarming security flaws in Apple's AirPlay protocol enable hackers to hijack millions of AirPlay-enabled devices like speakers, TVs, and even vehicle head units, posing a widespread risk due to manufacturers' failure to patch vulnerable third-party products.