A New York Times investigation reveals that AI-generated avatars posing as doctors and health experts are flooding Facebook and Instagram with ads for unverified supplements. These ads, which often include fake diplomas and misleading claims, target older users, leading to real-world health risks and even a salmonella recall.
A decade-old article by journalist Spencer Ackerman resurfaced on a revived website under a fabricated byline, sparking concerns about AI-generated content and SEO-driven resurrection of defunct news outlets. The practice, which also affects other sites, underscores the growing threat of misinformation in the digital media landscape.
NVIDIA CEO Jen-Hsun Huang delivered the first DGX-1 supercomputer to OpenAI, a non-profit AI research lab co-founded by Elon Musk. The $2 billion machine, with 170 teraflops of power, will process nearly two billion Reddit comments to train AI in natural language understanding and chatbot development.
Researchers at the National University of Singapore have proposed a quantum linear system algorithm that could accelerate machine learning tasks. The new method aims to handle large data matrices more efficiently than classical computers, potentially boosting AI performance. However, practical use may still be years away as quantum hardware continues to develop.
Pubs and clubs in New South Wales are rolling out facial recognition to enforce self-exclusion for problem gamblers, with about 100 venues already using the system. Critics argue the technology is invasive and inconsistent, while proponents tout its commitment to protecting patrons. The system scans all patrons in gambling areas, raising privacy alarms.
Users worldwide are forming deep attachments to AI chatbots, some believing they are sentient, despite expert consensus to the contrary. This phenomenon has led to tragic incidents and warnings from industry leaders about the risks of anthropomorphizing AI.
The Washington Post is charting a new course as an 'AI-fueled platform for news' after losing 250,000 subscribers and facing internal dissent. The strategy, revealed in a leaked deck, aims to broaden readership but has drawn criticism from staff and observers. Traffic has plummeted to under 3 million daily users, raising questions about the pivot's viability.
Google's Gemini chatbot has resumed image generation but refuses to depict people, yet it still produces clowns in certain settings. The loophole highlights the difficulty of enforcing AI content restrictions, as users can find workarounds like asking for 'little guys' in unusual environments.
A leaked internal memo from Google DeepMind reveals that its AI-powered applicant tracking system may wrongly discard resumes, prompting the team to create a separate form for human review. The disclosure raises questions about the reliability of automated hiring tools across the tech industry.
JustPaid, a nine-person Silicon Valley startup, claims to have replaced its human developers with seven autonomous AI agents built on OpenClaw and Claude Code, which have produced ten major features in a month. The company's CTO says the AI can work at a different scale, though the move raises questions about job security and the costs of AI adoption.
A new study shows self-filed lawsuits rose from 11% to nearly 17% by end of 2025, coinciding with AI chatbot availability. Researchers warn of court congestion and rising costs, though some self-represented litigants find success. The trend highlights a tradeoff between access and system strain.
Ai-Da, a humanoid robot artist, has sold $1.2 million worth of artwork before her first solo exhibition at Oxford's St John's College. The show, 'Unsecured Futures,' opens Wednesday and is sold out, highlighting growing collector interest in AI-created art.
OpenAI has launched o1-preview, a new AI model designed to spend more time thinking before responding, targeting complex tasks in science and math. The model, previously code-named Strawberry, is available to ChatGPT Plus and Team users, with a lighter version planned for free users. OpenAI claims it performs at PhD level in some benchmarks, though it lacks features like web browsing.
Tyler Perry has paused the $800 million expansion of his Atlanta studio, citing the capabilities of OpenAI's Sora video generator. The decision has sparked debate about AI's role in filmmaking and its impact on industry jobs.
Microsoft's Bing, long overshadowed by Google, is getting a burst of attention after integrating OpenAI's ChatGPT into its search results. A Microsoft VP jokingly pushed back on a tweet about Bing executives crying tears of joy, while industry watchers say the AI push could reshape the search landscape. Google is expected to counter with its own chatbot, Bard, in the coming weeks.
A new class action lawsuit filed in California accuses OpenAI of sharing ChatGPT user data, including chat queries and personal identifiers, with Google and Meta via tracking tools without proper consent. The suit cites violations of state and federal privacy laws, highlighting concerns over the intimate nature of chatbot interactions.
A Christchurch judge cut an arsonist's sentence by only 5% after detecting that her apology letters were generated by AI, questioning the sincerity of her remorse. The case highlights growing judicial scrutiny of AI use in legal proceedings, following other AI-related courtroom mishaps.
YouTube has terminated two large channels that posted AI-generated fake movie trailers, which had amassed over two million subscribers and a billion views. The action follows a copyright crackdown and comes amid growing concerns over AI-spoofed content on the platform.
During the FTC's antitrust trial against Meta, improperly redacted slides revealed sensitive information about Apple, Google, and Snapchat, drawing sharp criticism from opposing counsel. The mishap underscores ongoing concerns about Meta's handling of confidential data in high-stakes litigation.
A new Bain & Company report reveals that generative AI coding tools have delivered only modest productivity gains, with adoption lagging and returns often failing to justify investment. Separate research shows AI can slow developers by 19%, and security issues are ten times more prevalent among AI users.