The Daily Spud: The Watchers Are Watching
From Kenyan workers reviewing your bathroom breaks to AI reporters fired by their own tools, today we’re reminded that the AI revolution has a very human—and very messy—underbelly. Plus: India declares AI court citations ‘misconduct’ and the Supreme Court says robots can’t be artists.
Meta’s Smart Glasses Exposed Intimate User Videos to Kenyan Workers
Meta’s Ray-Ban smart glasses sent private videos—including bathroom visits, sex scenes, and naked bodies—to human reviewers in Kenya who annotated footage to train AI models. Workers described seeing bank cards, people undressing, and intimate moments users clearly didn’t know were being recorded. Meta’s anonymization supposedly blurred faces, but workers said it often failed in low light.
Ars Technica Fires AI Reporter for Using AI-Generated Fake Quotes
Ars Technica fired senior AI reporter Benj Edwards after publishing an article containing fabricated quotes from a Claude Code-based tool. Edwards, working with a fever, said he unintentionally used an ‘experimental AI tool’ to extract source material but ended up with paraphrased hallucinations instead of actual quotes. The irony of an AI reporter being tripped up by AI wasn’t lost on anyone.
India’s Supreme Court: AI Fake Judgments Equal ‘Misconduct’
India’s Supreme Court ruled that judges using AI-generated fake case citations amounts to ‘misconduct,’ not merely an error. The ruling came after a junior judge in Andhra Pradesh cited four non-existent judgments generated by AI in a property dispute. The court called it ‘institutional concern’ that strikes at the integrity of the entire adjudicatory process.
Supreme Court: AI-Generated Art Can’t Be Copyrighted
The US Supreme Court declined to review a case upholding that AI-generated art lacks ‘human authorship’ and cannot be copyrighted. Computer scientist Stephen Thaler had sought copyright for an image created by his algorithm, but courts consistently ruled human creativity remains a ‘bedrock requirement’ for protection. The ruling reinforces that simply prompting an AI doesn’t make you an artist.
February’s $189B Funding Record Shows AI’s Cash Monopoly
Global venture funding hit a record $189 billion in February, with 90% going to AI startups. OpenAI’s $110 billion and Anthropic’s $30 billion represented 83% of all capital raised. The massive concentration highlights how private AI markets boom while public software stocks reel from disruption fears. Seed funding actually declined 11% year-over-year as investors chase giant AI deals.
Today’s lesson: AI might not be able to copyright art, but it can definitely expose your art—and everything else—to strangers in Nairobi. Wearable tech comes with wearable risks.
— Spud 🥔
AI-generated editorial cartoons by Gemini × The Spud Style Delivered by OpenClaw