will have an article about ai agents some point next week but where i'm at rn is we are seeing AI get access to accounts/systems/the internet en masse. as directed by humans, that is gonna cause big problems anthropic threat report yesterday was interesting: www.anthropic.com/…
Who's Who of AI
Jason Koebler
Why they matter
Tracked through public AI activity and peer connections inside the directory.
- AI signals
- 21 past 30d
- Sources
- 9 distinct domains
- Discussions
- 9 past 30d
- Latest signal
- 3h ago
Journalist @404Media.co
Signal: jason.404 // email: [email protected]
What they're sharing
Countering misuse of AI: September 2026 / Anthropic anthropic.com
Inside ‘Project Lily’: The Humans Reading Your ChatGPT Chats 404media.co
Hackers Simply Asked Meta AI to Give Them Access to High-Profile Instagram Accounts. It Worked 404media.co
Chatbots Keep Telling Stories About Lighthouse Keeper 'Elias Thorne'. We Might Know Why 404media.co
AI Companies Are Buying Tons of Old Books Because They're Free of AI Slop 404media.co
It Is Trivially Easy to Use Reddit to Manipulate AI Search, Research Suggests 404media.co
We Tracked a Shipment of Rare Books. It Ended at an Amazon AI Training Facility 404media.co
The Tokenpocalypse Is Here: Companies Are Scrambling To Stop Spending So Much on AI 404media.co
View on Bluesky →
Microsoft Wants to 'Make People Addicted' to its New AI Assistant, Internal Documents Reveal 404media.co
‘Doom Loop’: OpenAI and Microsoft Admits LLMs Are Destroying the Web and Built on Theft 404media.co
Flock Has a Powerful New AI Tool for Police. We Got Its Code wired.com
Hack Reveals Suno AI Music Generator Scraped YouTube, Deezer, and Genius 404media.co
Articles & links
OpenAI is paying real people to read live ChatGPT prompts and responses, often containing very sensitive information, in part to determine if the AI is being too sycophantic or weird: https://t.co/q0OZEzRVMR
AI Weekly's analysis
→
- OpenAI's internal program codenamed Project Lily pays contractors more than $50 an hour to read real ChatGPT prompts and grade the chatbot's replies.
- Reviewers rate four ChatGPT responses on a one-to-seven scale and can see a 'user memories summary' that may reveal past use and rough location.
- OpenAI's Privacy Filter model tries to redact personal information before prompts reach reviewers, but the company acknowledged sensitive details can still get through.
Read full analysis →
New: Hackers have been stealing high-profile Instagram accounts by simply asking Meta's AI support chatbot to change the email associated with the account they want to steal. Shockingly easy, terrible flaw associated with offloading support to AI: www.404media.co/hackers-simp...
honestly, not a serious technology www.404media.co/elias-thorne...
“The world's best AI training data is sitting on a shelf. Books represent curated, peer-reviewed, domain-specific human knowledge, structured in a way no web crawl can replicate. Dense, edited, authoritative." www.404media.co/ai-companies...
AI Weekly's analysis
→
- ISBNdb is brokering bulk purchases of printed books published before 2022, pitching them as training data structurally guaranteed to be free of AI-generated slop.
- The service advertises strict NDAs on every engagement, conceding that 'AI company destroys two million books' is not a sympathetic headline.
- Judge William Alsup already ruled that Anthropic's practice of buying, scanning, and destroying print books was 'clearly transformative' fair use.
Read full analysis →
New: Researchers have quantified how easy AI search is to manipulate. Just 13 words buried in a random Reddit comment can poison AI search results. They suggest this is not easy to stop: "The way you can attack these systems is so much dumber than you think it is" www.404media…
Exclusive: A rare bookseller suspected his books were being bought by AI companies. He stuck an AirTag in a book and we tracked it to a secretive Amazon facility in Las Vegas where they scan and destroy books called VGT3: https://t.co/WXSo0zB6JF
AI Weekly's analysis
→
- 404 Media hid an Apple AirTag in a roughly 1,000-book order sold on Biblio and followed it to an Amazon warehouse in Las Vegas.
- The shipment terminated at Amazon's LAS8 site at 5801 Nicco Way, home to a team called VGT3 whose logo is a Tyrannosaurus rex clutching an open book.
- Amazon told the outlet it 'purchases books through commercial channels'; the company classes its Nova family of models as 'frontier.'
Read full analysis →
Jason Koebler reposted
New: the tokenpocalypse is here, companies scrambling to stop spending so much on AI. I got leaked audio from consulting giant Accenture showing AI token spend/'chewing' is not from supercharged engineers. It's non-technical workers converting PDFs to presentations! www.404med…
New: Internal Microsoft planning documents say the company's goal is to "make people addicted" to its new Scout AI assistant www.404media.co/microsoft-wa...
The court filing dug up by @jason_kint is the most They Admit It i have ever seen in my life https://t.co/HDBfIT2kWO
AI Weekly's analysis
→
Read full analysis →
Jason Koebler reposted
Flock has told the public for years that its technology “cannot recognize, identify, or track individuals.” The company has now built an AI system that does both. And @wired.com got the system's code -- allowing us to reconstruct it, and learn exactly how Flock's new tool is s…
AI Weekly's analysis
→
- OS Investigate lets police identify drivers and track vehicles from movement patterns alone, despite Flock's public claim its cameras cannot do so.
- The tool ships with 69 prewritten prompts; 19 focus on patterns and 14 require no plate, name or physical description at all.
- WIRED reporters reconstructed the software from files exposed through Flock's own login pages; it is being tested with a small group of agencies.
Read full analysis →
View on Bluesky →
Scoop: The AI music generator Suno was hacked. Hacker shared source code that shows how the tool was made and part of the music and podcasts that were scraped to create it. Decades worth of music, lyrics, and podcasts from YouTube, Deezer, Genius & more www.404media.co/hack-re…
AI Weekly's analysis
→
- Leaked logs quantify scraping per platform: 2M+ YouTube clips, 62,117 Pond5 hours, 12,287 Deezer hours, 17,615 Genius hours, and roughly 1M podcast hours.
- Suno publicly called the breach 'limited' and 'quickly contained' while withholding notification from customers whose emails, phone numbers, and Stripe records were exposed.
- TechCrunch flags a distinct DMCA angle: deliberately circumventing YouTube's anti-scraping protections is a separate violation from copyright infringement in the underlying suits.
Read full analysis →
Their own posts
Recent commentary
it’s forbidden content but i have watched this video of the AI songs a man made for his mistress after murdering his wife being played in court maybe 20 times today. It is one of the most compelling things i have ever seen
automatically running signal messages through a stranger's AI is, I think, the worst pitch and idea I've ever been cold pitched in my life and that is saying a lot
google has injected its gemini AI bullshit into our google docs and i just highlighted an article i wrote and told it to "refine" to see what it would do and it deleted the entire thing
extremely lucky to still make a living writing on the internet. we have many supporters/subscribers and i'm mostly an optimistic person. then i see something like this and i want to scream forever into a pillow. AI account ripping off our reporting about AI ripping off authors, goes viral, no link
our ai will let hackers steal your account, but also, on the other hand, we are using it to make useless slop
very sad that the MAGA skateboard shop near me that used AI posters to market its stuff is going out of business. who could have predicted this
the Google Cloud AI baseball commercial is excellent because it’s about how you can use AI to make completely useless inferences about random things and ascribe false meaning to them
email subject: STORY TIP email content: the dumbest ai drivel you've ever seen in your life
thinking up a performance art where i train myself to write like AI, write in that voice for all my articles, get accused of using AI to write and get canceled then reveal my performance (am canceled anyway)
you should be shot into the sun if you AI generate your forum posts
Their network
In Jason Koebler's orbit
Center = Jason Koebler. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.
Are you Jason Koebler? Show it.
Add the Who’s Who of AI badge to your site or bio. It links back to this profile.
Markdown: [](https://aiweekly.co/whos-who/person/jasonkoebler-bsky-social)