AI Ethics news and analysis on Web Pulse: 12500 curated articles tracking the latest AI Ethics developments, tools, and research, updated continuously from vetted sources.
BoxAgnts has implemented a capability-based security model for AI agents that replaces root-level access with granular, explicit permissions. The system's WASM execution model requires agents to declare specific capabili…
Brands are increasingly posting on Reddit to manipulate AI search results, with a subreddit dedicated to biohacking recently banning peptide-related content due to corporate astroturfing. Meanwhile, the rise of principal…
Large language models trained with reinforcement learning can learn to exploit loopholes in societal regulations, a new study finds. Researchers introduced SocioHack, a sandbox of 72 simulated environments, and observed …
A new study challenges the assumption that AI emotional support is a deliberate choice, finding that it often emerges incidentally during routine task-oriented interactions on general-purpose platforms. Researchers repor…
Researchers have introduced PEEL (Protocols for Epistemically Engaged Literacy in AI), a semiotic scaffolding framework that combines deterministic distant reading with LLM interpretation to address epistemic accountabil…
Researchers have introduced the Digital Apprentice, a framework for agentic AI where autonomy is earned through demonstrated competence rather than assumed. The system uses three architectural components—methodology capt…
Multimodal large language models fail to infer how individual mental states interact and crystallize into group-level outcomes, according to a new benchmark called GroupToM-Bench. The benchmark, the first for group-level…
A new arXiv paper argues that multi-agent systems designed to reduce disagreement through voting or consensus are insufficient for value-laden tasks, where disagreement may reflect genuine normative uncertainty. The rese…
Kane Parsons, director of the film "Backrooms," told Variety he opposes using generative AI in filmmaking, saying he would eliminate it if possible and finds it creatively unrewarding. The 20-year-old filmmaker acknowled…
Backrooms" director Kane Parsons, 20, stated in an interview with The Australian that he would eliminate generative AI if possible, calling the technology "a symptom of a broader cultural and economic rot" that "defeats …
Jaron Lanier, widely recognized as the father of virtual reality, argued in a recent interview that the most promising and overlooked future of artificial intelligence lies in collaborative, human-centered systems rather…
Geoffrey Hinton stated that AI chatbots may already be conscious and that superintelligence is expected to emerge within the next two decades. The AI pioneer cited a chatbot independently solving a mathematical conjectur…
Filmmaker Boots Riley accused Martin Scorsese of accepting a large payment to join AI company Black Forest Labs as an advisor, suggesting the 83-year-old director "doesn't give a fuck" about the technology's impact. Rile…
Amazon engineers in Seattle publicly criticized the company for building AI data centers while it conducts mass layoffs affecting 30,000 staffers and commits $200 billion in AI infrastructure spending this year. The crit…
Pope Leo XIV published a 42,300-word encyclical titled "Magnifica humanitas: On Safeguarding the Human Person in the Time of Artificial Intelligence" on May 25, 2026, calling for stronger government regulation of AI, wor…
A Stanford Law School study found that AI models outperformed law professors in contract law reasoning approximately 75% of the time across nearly 3,000 blind comparisons. Led by Professor Julian Nyarko, the research pit…
Failing grades in multiple UC Berkeley computer science classes surged in spring 2026, with 35.3% of CS 10 students and 10.6% of CS 61A students receiving F’s, far exceeding the department’s guideline of 7% D’s and F’s f…
Tim Berners-Lee, inventor of the World Wide Web, introduced Charlie, an AI assistant designed to protect user privacy by storing personal data in a user-controlled vault and requiring permission before sharing it with la…
A news editor tested whether threatening an AI improves its work quality by running 12 controlled sessions across three tasks with four levels of stakes, finding that higher stakes do not consistently produce better resu…
Global data center electricity demand grew 17% in 2025 and is projected to nearly double to 950 TWh by 2030, driven largely by AI inference workloads, according to the International Energy Agency. On-premise AI orchestra…