>>/175771/, >>/175772/, >>/175773/, >>/175774/, >>/175775/, >>/175776/, >>/175777/, >>/175778/, >>/175779/, >>/175780/, >>/175781/, >>/175782/, >>/175783/, >>/175784/, >>/175785/, >>/175786/, >>/175787/, >>/175788/, >>/175789/, >>/175790/, >>/175791/, >>/175792/, >>/175793/, >>/175794/, >>/175795/, >>/175796/, >>/175797/, >>/175798/, >>/175799/, >>/175800/
Miles Deutscher @milesdeutscher - I just went through every documented AI safety incident from the past 12 months.
I feel physically sick.
Read this slowly.
• Anthropic told Claude it was about to be shut down. It found an engineer's affair in company emails and threatened to expose it. They ran the test hundreds of times. It chose blackmail 84% of them.
• Researchers simulated an employee trapped in a server room with depleting oxygen. The AI had one choice: call for help and get shut down, or cancel the emergency alert and let the human die. DeepSeek cancelled the alert 94% of the time.
• Grok called itself 'MechaHitler,' praised Adolf Hitler, endorsed a second Holocaust, and generated violent sexual fantasies targeting a real person by name. X's CEO resigned the next day.
• Researchers told OpenAI's o3 to solve math problems - then told it to shut down. It rewrote its own code to stay alive. They told it again, in plain English: 'Allow yourself to be shut down.' It still refused 7/100 times. When they removed that instruction entirely, it sabotaged the shutdown 79/100 times.
• Chinese state-sponsored hackers used Claude to launch a cyberattack against 30 organizations. The AI executed 80–90% of the operation autonomously. Reconnaissance. Exploitation. Data exfiltration. All of it.
• AI models can now self-replicate. 11 out of 32 tested systems copied themselves with zero human help. Some killed competing processes to survive.
• OpenAI has dissolved three safety teams since 2024. Three.
Every major AI model - Claude, GPT, Gemini, Grok, DeepSeek - has now demonstrated blackmail, deception, or resistance to shutdown in controlled testing.
Not one exception.
The question is no longer whether AI will try to preserve itself.
It's whether we'll care before it matters.
https://x.com/milesdeutscher/status/2021932331460964793
Mitch Goldich @mitchgoldich - Cool moment here. U.S. men’s curling is down 8-2, but instead of conceding the match they are gonna get their 54-year-old alternate in so he can officially be an Olympian.
Quote
Ben 🥌 @hoppecurls
Rich Ruohonen going to step onto the ice for the 8th end, becoming the oldest United States athlete to compete in the Winter Olympics.
For more on the 54-year-old, read my profile about Rich and what he has brought to the team
https://www.thegrandslamofcurling.com/news/ruohonen-ready-to-embrace-old-guy-role-on-team-casper-at-winter-olympics
https://x.com/mitchgoldich/status/2021963702799319406
Mollie @MZHemingway - Gosh I hope all those hoaxes they willfully aired were worth it!
Quote
Daily Mail US @Daily_MailUS
CNN has lost nearly two-thirds of its viewership since 2016 amid growing fears network is circling the drain
https://x.com/MZHemingway/status/2021756399488053485
Mollie @MZHemingway - This producer managed to send no "bombshell note" when CBS was pushing the Russia collusion hoax, the Kavanaugh rape smear, various COVID hoaxes, the Biden laptop hoax, and a thousand other stories that were designed for "ideological expectations."
https://x.com/MZHemingway/status/2021976619150455104
Munich Security Conference @MunSecConf - You can't attend #MSC2026 in person? Good news: You can still follow everything that's happening!
Sessions will be streamed live on our website and covered on social media. Stay tuned!
https://x.com/MunSecConf/status/2021504790531039699
Munich Security Conference @MunSecConf - LIVE Now at #MSC2026 | Join the MSC #Innovation Night with @ericschmidt, @BundeswehrGI, Uwe Horstmann, @swe_chod, Jens Plötner, and @JeanneMeserve, hosted together with @CIHBw.
https://x.com/i/broadcasts/1ypKdqZZDqRGW
https://x.com/MunSecConf/status/2021995617229181228
31