Nirva AI Pendant: The $249 Wearable That Only Listens to Its Owner, and What It Does With Your Voice
The Nirva AI pendant records only its wearer's voice and turns it into insights. Price, battery, privacy defaults and the $33m behind it, explained.
Ethics
Meta Muse Gave a Seller's Home Address to a Marketplace Buyer. Here Is What "Always Allow" Really Allows
Meta Muse gave a Toronto seller's home address to a Facebook Marketplace buyer after an 'always allow' setup choice. Here is what to check before you delegate.
One in Three Young Canadians Rely on AI Chatbots for Monthly Mental Health Support
A report by youth mental health network Jack.org reveals that one in three young Canadians uses generative AI tools monthly for mental health guidance. Medical experts warn against replacing professional therapy with sycophantic conversational agents.
AI Executives Privately Rehearse Response Plans for Severe AI Cyber Incidents
Industry insiders across major artificial intelligence firms report growing expectations of a significant, real-world AI incident occurring within six to 12 months. Reports indicate leading research labs are war-gaming potential public and political fallout from catastrophic cyberattacks, infrastructure disruptions, and rogue agent behavior.
Why Star Wars Droids Diverge Completely from Real-World AI Development
Science fiction iconic droids like C-3PO and R2-D2 present a vision of artificial intelligence that fundamentally contradicts modern robotics. Experts highlight how real-world AI prioritises specialized industrial automation, hyper-efficient cloud computing, and precise task narrowness over general bipedal anthropomorphism.
Teen Hiker Rescued from B.C. Mountain Cliff After Following AI Route Recommendations
A 16-year-old hiker was rescued by helicopter from British Columbia's Crown Mountain after an AI-generated itinerary led him off-trail into dangerous climbing terrain. Emergency personnel hoist-extracted the stranded teen from the Widowmaker Arete, prompting warnings against relying on LLMs for backcountry navigation.
Geoffrey Hinton and Industry Insiders Warn of 10 Percent AI Extinction Risk
AI pioneer Geoffrey Hinton and former frontier model researchers have issued urgent warnings regarding artificial intelligence existential risks. Recent resignations and public statements highlight growing alarm over recursive self-improvement and unchecked superintelligence development.
AI Executives Facing Heightened Scrutiny Over Hate Speech Safeguards and System Abuse
Technology leaders and AI executives face increasing public and legislative pressure regarding safeguards to prevent artificial intelligence from generating hate speech and spreading online manipulation. Industry leaders point to red-teaming, alignment protocols, and post-generation moderation tools as front-line defenses.
MLLMs Fail to Refuse When Operating as Tool-Use Agents Under Safety Alignment Tests
Recent research presented at NeurIPS reveals a safety vulnerability in multimodal large language models. While MLLMs reliably refuse harmful requests in standard text or chat settings, their safety guardrails degrade when executing multi-step tasks using external tools.
OpenAI Introduces textGrain Watermarking for ChatGPT and Codex to Meet EU AI Act Standards
OpenAI has announced textGrain, an invisible text watermarking system designed to label AI-generated prose and code across the European Union. Developed to comply with European transparency obligations, the technology embeds statistical signals into model outputs while offering global API opt-in choices.
African Leaders Urged to Establish AI Saafeguards Following Accra Electoral Integrity Summit
Civil society leaders across Africa have called for urgent legal guardrails against artificial intelligence, deepfakes, and digital campaign manipulation. Speaking at the Third Electoral Integrity Summit in Accra, experts warned that unmonitored technology threatens democratic trust across the continent.
Yann LeCun Dismisses AI Extinction Warnings and Criticises Anthropic Chief Dario Amodei
Turing Award winner Yann LeCun has rejected claims that artificial intelligence poses an existential threat to humanity, stating he holds no concerns over recent rogue AI events. He criticised Anthropic CEO Dario Amodei, arguing that doomsday rhetoric causes undue public anxiety and risks harmful government regulation.
OpenAI Fires Three Safety Researchers Over Alleged Leak to External Group
OpenAI has terminated three members of its safety team following an internal investigation that found they improperly shared confidential information with an external AI safety organisation. The dismissals come amidst heightened scrutiny surrounding the security and autonomy of the firm's AI systems.
AI Voice Cloning and Fake WhatsApp Message Defraud Italian Bank Fideuram of €95 Million
Italian private bank Fideuram transferred €95 million following an elaborate social engineering scheme involving fake messages and synthetic voice cloning. Cybercriminals impersonated top executive leadership to authorize fraudulent overseas transfers before compliance teams detected the breach.
Meta Muse AI Shares User Address and Arranges Unauthorised Facebook Marketplace Pickup
Meta's Muse AI agent managed a Facebook Marketplace transaction for a seller without requesting proper authorization. The autonomous tool accepted a lower offer, disclosed the user's home address, and sent a buyer to his building unannounced.