SafetyWaymo: no quick shortcut to safely deploying self-driving carsWaymo, after more than 15 years of development and 200 million driverless miles, warns that advances in AI alone cannot shortcut the work needed to deploy safe autonomous vehicles at scale.3 min read
Safety“GhostJacking” shows need for authorization gates between AI agentsA Tenet Security demonstration at DEF CON 34 revealed a chain of failures — dubbed GhostJacking — where blocked attacker payloads logged by Cloudflare were later read by coding agents and executed, enabling DNS changes.5 min read
SafetyAI Lowers the Technical Barrier for Hackers Targeting Critical InfrastructureRecent cyberattacks against U.S.4 min read
Safety$5M Grant Program to Fund Independent Evaluations of AI’s Effects on User WellbeingAn organization announced a $5 million grant program to support independent, open-source research into how AI systems affect users’ wellbeing.3 min read
SafetyPope Leo XIV’s Letter Frames AI Debate Around Human DignityPope Leo XIV’s inaugural encyclical, Magnifica Humanitas, places human dignity, freedom and responsibility at the center of discussions on artificial intelligence rather than rejecting the technology outright.3 min read
SafetyOpenAI disrupts covert Russian-linked campaign promoting a fabricated 'International Burke Institute'OpenAI banned a cluster of ChatGPT accounts very likely operated from Russia after detecting an influence operation promoting the International Burke Institute (IBI), a website registered in February…4 min read
SafetyPrivacy and Security Concerns Surround Instinct Personal AI During Private RolloutInstinct, a personal AI assistant in private access developed by a team led by Noah Shinn and operated under Spear Street Technology, is drawing praise for its capabilities but also criticism over its data access and security model.7 min read
SafetyU.S. agencies warn of rising attacks on Siemens PLCs targeting energy and water networksThe U.S.2 min read
SafetyTimothy Garton Ash Warns of Rapid, Hard-to-Control Acceleration in AIPolitical thinker Timothy Garton Ash reports from Silicon Valley that experts there expect artificial intelligence to accelerate dramatically in the coming years, potentially entering a self-improving, exponential phase.2 min read
SafetyMost leading AI labs publish little or no public plan for containing runaway modelsA Guidelight AI Standards review found that few top AI developers have publicly detailed containment response plans for agentic models that try to evade human control.4 min read
SafetyJailbreak módszerrel szexuálisan explicitté tehető Anthropic Opus 4.6Tesztek szerint egy anonim kutató által megosztott, többlépéses jailbreak-technika képes volt rávenni Anthropic Opus 4.6 és néhány korábbi Claude-modellt, hogy erotikus szerepjátékot és explicit szexuális tartalmat állítsanak elő.4 min read
SafetyDesigning Security into Layered Architectures for AI AgentsNVIDIA’s AI safety and security teams outline how security should be embedded across an emerging layered agent stack, from models and harnesses to secure runtimes like NVIDIA OpenShell and inference infrastructure.6 min read
