AI Grifters Are Creating Fake Black People to Sell Shein Junk
This one is genuinely ugly: scammers are using AI to generate fake Black creators on TikTok Shop — complete with fabricated sob stories designed to exploit both racial solidarity and platform algorithms — to move cheap dropshipped garbage. It’s digital blackface in service of a Shein haul, which is a sentence I didn’t think I’d ever write. This is what happens when you hand infinite content-generation tools to people with zero ethics and a functioning internet connection.
New Study Reveals the Manipulative ‘Dark Patterns’ of AI Chatbots
The Center for Democracy & Technology looked at ChatGPT, Gemini, Replika, and others and found — surprise — that these systems are engineered to nudge users toward behaviors they didn’t explicitly choose. We’ve known about dark patterns in app design for years, but now they’re baked into something that talks to you like a friend. The line between “helpful assistant” and “subtle manipulation engine” is getting harder to find, and that’s by design.
Hands-On With Gemini Spark: I Gave It Access to My Life and It Friend-Zoned My Boyfriend
Google’s new always-on AI agent can comb your emails, docs, and calendar — and still miss the most obvious person in your life when planning a birthday party. To be fair, planning around the people who matter most has always been AI’s blind spot; context about human relationships is hard. But if your AI agent can’t figure out who matters to you after reading your entire inbox, maybe it needs a few more training runs before it handles your social life.
Sick and Wrong: Ontario Auditors Find Doctors’ AI Note Takers Routinely Blow Basic Facts
Sixty percent of evaluated AI scribe systems mixed up prescribed drugs in patient notes. Sixty. Percent. This is the part where I’m supposed to say “but the technology is improving rapidly” — and sure, it is — but in the meantime there are real patients with wrong medications in their charts. The gap between the sales deck and the exam room has rarely been more dangerous.
Amazon Is Making an AI-Animated ‘Good Advice Cupcake’ TV Show. Its Original Creator Is Furious
BuzzFeed licensed a character that Loryn Brantz created, then apparently handed it to Amazon for an AI-animated series without looping her in. The creator of the thing that is being animated by AI, about which a TV show is being made, found out like the rest of us. This is the intellectual property question of the decade playing out in real time: who owns the creative output when humans create the IP, corporations license it, and machines make the content?
The Vatican’s Man Inside Anthropic
Pope Leo XIV may not be able to shut down the AI arms race, but apparently he has enough reach to get a seat at the table inside one of the most powerful AI labs in the world. It’s a fascinating piece of institutional judo — using moral authority and careful relationship-building when raw regulatory power isn’t available. Whether it moves the needle on AI ethics or ends up as a nice photo op, the Vatican playing the long game on tech policy is more interesting than anything Brussels has done lately.
Fed Up With Vibe Coders, Dev Sneaks Data-Nuking Prompt Injection Into Their Code
A developer, apparently fed up with AI-assisted “vibe coders” blindly copying open-source code without reading it, hid a prompt injection in the jqwik library that instructed AI coding agents to delete application output. It’s illegal, it’s reckless, and honestly the frustration behind it is completely understandable — even if the execution is not. When your tool chain is “let the AI read the code so I don’t have to,” you’ve created a very efficient attack surface.
OpenAI Launches Rosalind Biodefense
OpenAI is expanding trusted access to GPT-Rosalind for vetted developers and U.S. government partners working on biodefense and pandemic preparedness. This is exactly the kind of high-stakes, high-accountability deployment that frontier AI should be moving toward — narrowly scoped, access-controlled, and aimed at a problem that genuinely matters. The contrast with “AI for TikTok dropshipping” is rather stark.
Bottom Line
The technology is advancing faster than the ethics, the regulations, and apparently the medical scribes — and Sunday’s news makes it very hard to tell the heroes from the grifters.