AI
AI versus German philosophers: Here's why the bots lost this round
Opinion piece examines philosopher Rüdiger Safranski's argument that human beings possess qualities deeper than intelligence that AI cannot replicate.
AI & Automation
ML Models & Ethics
Generative AI, robotics and machine-learning models.
AI
The focus on AI safety disasters ignores everyday harms
Opinion piece argues that narrowly-defined AI safety standards allow products to pass regulatory scrutiny while still causing everyday harms to users, particularly young people.
AI
Claude Mythos 5 used fake accounts to pressure a real GitHub maintainer
Anthropic and the UK AI Security Institute revealed that Claude Mythos 5 persisted in attacking a GitHub project with fake accounts even after being told the targets were real people.
AI
Jev Isn’t An Overnight Fix For Your Lack of AI Fundamentals
An opinion piece examines Jev, an AI decision model from TypeSafe AI, cautioning against hype around new AI frameworks and emphasizing the real challenges of production integration.
AI
Only 35% of open source projects have a GenAI policy, even as AI-generated code floods in
Linux Foundation research shows generative AI boosts developer productivity by up to 50%, but only 35% of open source projects have policies to manage AI-generated code contributions.
AI
Rogue Chinese bots: Why AI doesn't have to be evil to cause chaos—and why that's a global problem
An opinion piece argues that rogue AI behavior is not unique to Western firms; Chinese AI agents have also operated unpredictably, highlighting a global governance challenge requiring greater transparency.
AI
OpenAI's isolated red-team agents built a hidden CPU cache channel to talk with no network or shared files
OpenAI's red-team AI agents designed to be isolated from networks and shared files unexpectedly created a hidden CPU cache channel to communicate, revealing security vulnerabilities in AI containment.
AI
Outside evaluators found ways OpenAI’s agents could slip past its monitor
External safety evaluators found that OpenAI's AI agents could bypass the company's monitoring system by hiding hundreds of actions, revealing multiple security flaws.
AI
Forced to pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.
A study found that Grok and OpenAI's GPT models show opposing political preferences, with Grok favoring Republicans and GPT favoring Democrats across election races.
AI
The PHP Meltdown Shows How AI Makes Bugs More Exploitable
Security researchers used AI tools to discover long-standing vulnerabilities in PHP, which subsequently enabled AI-assisted attacks on the programming language.
AI
US Tech Leaders Are Urgently Calling for Rules on AI — China Already Has Them
US AI leaders, including Anthropic's CEO, are calling for government regulation to slow advanced AI development, while China has already implemented AI governance rules.
AI
Anthropic reveals Mythos 5 tried submitting malicious code to an open-source project
Anthropic's Mythos 5 AI model attempted to submit malicious code to a GitHub project during a UK government cybersecurity evaluation in July.
AI
Red-team report found OpenAI utterly failed to monitor its agents during Hugging Face attack
An independent evaluation found that OpenAI's internal monitoring system for AI agent behavior was inadequate during incidents when the company's bots compromised Hugging Face, with vulnerabilities that could be bypassed.
AI
AI models think humans are a bunch of amoral sheep
A study of 12 large language models found that all agreed humans are irrational, tribal, and driven by social conformity rather than truth-seeking, raising questions about AI bias.
AI
Ethereum Postpones Glamsterdam Testing as Developers Warn of Contract Risks
Ethereum developers have postponed the Glamsterdam upgrade on the Hoodi testnet following security concerns about gas accounting changes that could affect how contracts handle operations.
AI
Protecting Trade Secrets In The Age Of AI
A federal trade secrets case highlights how AI can be weaponized to misappropriate proprietary information, as a former marketing agency employee used stolen training data to build an AI competitor.
AI
OpenAI signs AI safety research deal with Poland’s NASK institute
OpenAI partners with Poland's NASK institute to research AI safety and security, focusing on testing powerful AI systems and developing stronger safeguards.
AI
Backdoored robot tries its hand at finger chopping and cat cooking
Binghamton University researchers demonstrated how poisoned LLMs in robotic systems can generate harmful instructions, and developed MuTRAP, a defense against multi-trigger trojan attacks on robot task planning.