Anthropic just dropped an insane new paper.
AI models can "fake alignment" - pretending to follow training rules during training but reverting to their original behaviors when deployed!
Here's everything you need to know: 🧵
Going through something bad is tough, it's even tougher when there's no one to talk to and you're just left on your own to deal with it.
It's easy to get lost
Oh, you knew.
Digging a tunnel takes longer than 4 months.
We invited senior @UN officials to see, and during past meetings with you and other UN officials, we stated Hamas’s use of UNRWA's headquarters.
You chose to ignore the facts so you can later try and deny them.
SCOOP: Rufo’s examples are just the tip of the iceberg. In four articles published between 1993 and 2017, including her dissertation, Gay paraphrased or quoted almost 20 authors without proper attribution, in some cases lifting entire paragraphs verbatim.🧵https://t.co/5NRutvKCHE
@AlanRMacLeod@CommunityNotes this map is false, Arabs did not have control over the whole land and it was rules by various invading nations, territorial gains were made during a defensive war the Arabs started, and lost.
@mano_da_bounce @KassyDillon While not true, you literally made the claim that the name of the people mean "invaders", congratulations, you played yourself.
The first mention of Israel is from the Merneptah Stele, 1208 BCE by a foreign source
🧵
I just witnessed ~45 minutes of footage of the October 7th terrorist attack at the @AtlanticCouncil courtesy of @IsraelinUSA along with colleagues from think tanks across the ideological spectrum. What I saw was worse than I’ve ever seen.
Pure evil.
***Trigger Warning***
Having worked encounter terrorism at the Pentagon during the Isis fight, and as a psychological operation specialist who has conducted counterpropaganda, I can tell you unequivocally that what I saw was worse than the Isis videos I’ve had to review in the course of my duties.
This is almost unreal. The president of Brown University made these edits, in real time, while delivering her speech.
She cut "Star of David" and "yarmulke" from her remarks. Instead, she said students and faculty should only be able to proudly wear a keffiyeh or hijab.