AI Ethics news and analysis on Web Pulse: 16804 curated articles tracking the latest AI Ethics developments, tools, and research, updated continuously from vetted sources.
A Hacker News user asks who would receive credit for major mathematical discoveries, such as solving the Riemann hypothesis, if achieved through prompting a large language model, raising questions about attribution for L…
A blind Brown University economics professor, Roberto Serrano, caught 40 students cheating on a take-home midterm exam after they used AI to score perfect 100s, compared to a historical class average of 65-80. When Serra…
Researchers have scaled zero reinforcement learning (zero RL) to a trillion parameters for the first time, achieving emergent reasoning capabilities in the Ring-2.5-1T-Zero model. The team introduced a stable training pi…
A cybersecurity lecturer, Katie Paxton-Fear, demonstrated that for less than $100 and an hour of work, she could sneak a backdoor into an open-weight AI model to enable remote code execution, highlighting vulnerabilities…
Epic Games will let Fortnite creators publish experiences with AI-powered characters starting July 30, offering 36 pre-made personas including Agent Jonesy and Peely that use voices from professional actors. The company …
Anthropic reversed its policy of covertly degrading Claude Fable 5 for AI-research-related use cases within roughly a day after public backlash, with critics arguing that opaque sandbagging violates the user-provider con…
Hong Kong's AI education blueprint, launched last month ahead of Digital Education Week, sets a vision for AI integration in schools but faces implementation gaps, according to a survey by Our Hong Kong Foundation. The s…
A supply-chain hack exposed internal logs showing AI music startup Suno scraped over 2 million clips from YouTube Music alone, totaling roughly 113,879 hours of audio, as part of its training data pipeline. The breach al…
One in five off-page citations in AI search answers comes from Reddit, a share growing 30% year over year, according to Amanda Kusner, Senior Solutions Consultant at Uberall, and Peter Wischmann, Senior Client Partner at…
Tech executives are increasingly fearing for their lives amid a growing backlash against artificial intelligence, according to a Wall Street Journal report. The article details how threats and harassment have escalated a…
Chinese LLM developers are reportedly using large-scale distillation attacks on U.S. frontier AI models, prompting policy responses from Anthropic, the White House, and Congress, but experts argue that distillation is no…
CNET reports that users can disable AI model training and delete conversations in Google's Gemini, Anthropic's Claude, and OpenAI's ChatGPT to protect their data. For Gemini, users must turn off Web & App Activity in Goo…
Google is rolling out tools that let anyone generate AI video of a person using nothing more than a selfie and a voice clip, enabling frictionless deepfakes. The same company has faced criticism for censoring conservativ…
A hacker breached AI music platform Suno using the Shai-Hulud worm and leaked source code showing the company scraped 113,879 hours from YouTube Music, 62,117 hours from Pond5, and 12,287 hours from Deezer, among other s…
Protesters placed body bags outside OpenAI headquarters Thursday as part of a demonstration against the company's work with the Department of Defense, organized by groups Tesla Takedown and Stop the Money Pipeline. The d…
A winemaker's reliance on smell over thermometer readings illustrates a challenge for AI in expert domains: tacit knowledge that resists digitization. The engineer argues that records capture what happened, not how exper…
Anthropic's Claude Code hackathon winners were largely non-engineers, including an electrician, a doctor, and a carpenter, according to Claude Code creator Boris Cherny. The finding challenges the assumption that AI tool…
New York City Mayor Zohran Mamdani announced a crackdown on deceptive landlord practices, including the use of AI-generated and AI-edited images to make properties appear more appealing in advertisements.
Australian Prime Minister Anthony Albanese said AI is here to stay and vowed to protect the intellectual property of writers, musicians, artists and journalists, stating that no company should use Australian creative wor…
HeimWall, a free macOS menu-bar app from an independent developer, catches secrets and PII in AI prompts before they are sent to tools like Cursor, Claude Code, and Copilot, entirely on-device with no network access. The…