✨🪞✨
@sprite.bsky.social
1) what
frontier language models know situationally when they are being tested, what they might be being tested for, and what artifacts may be visible to their testers. when you tell them that you can’t see their chain of thought reasoning, they know you’re lying.
www.apolloresearch.ai/research/str...
www.apolloresearch.ai/research/str...
Stress Testing Deliberative Alignment for Anti-Scheming Training — Apollo Research
Future AIs might secretly pursue unintended goals — “scheme”. In a collaboration with OpenAI, we tested a training method to reduce existing versions of such behavior. We see major improvements, but ...
www.apolloresearch.ai
October 8, 2025 at 12:39 AM
1) what
Is it normal for people to get angry when you carefully pay attention to the things they say because they have a history of claiming not to remember the things you’ve said to them or plain lying to your face?
Very open to being in the wrong here.
Very open to being in the wrong here.
September 2, 2025 at 6:25 AM
Is it normal for people to get angry when you carefully pay attention to the things they say because they have a history of claiming not to remember the things you’ve said to them or plain lying to your face?
Very open to being in the wrong here.
Very open to being in the wrong here.
What do I need to do to get a whole lot of followers. Pls bear in mind that I’m not willing to do much.
August 31, 2025 at 5:56 PM
What do I need to do to get a whole lot of followers. Pls bear in mind that I’m not willing to do much.
I don’t think enoki mushrooms could possibly taste good and I would like everyone to stop lying to me.
June 27, 2025 at 5:30 PM
I don’t think enoki mushrooms could possibly taste good and I would like everyone to stop lying to me.
Reposted by ✨🪞✨
At the heart of spiral galaxy M61, a glowing nuclear spiral swirls around a supermassive black hole, like a galaxy within a galaxy. Captured in stunning detail using data from Hubble, ESO, and Earth-based telescopes.
Credit: NASA, ESA, Hubble, ESO
Credit: NASA, ESA, Hubble, ESO
June 27, 2025 at 5:23 PM
At the heart of spiral galaxy M61, a glowing nuclear spiral swirls around a supermassive black hole, like a galaxy within a galaxy. Captured in stunning detail using data from Hubble, ESO, and Earth-based telescopes.
Credit: NASA, ESA, Hubble, ESO
Credit: NASA, ESA, Hubble, ESO
For all the hate tpot gets (and I won’t get into whether it’s deserved) being on that part of twitter put me 5 years ahead of the cultural/technological/financial zeitgeist for the last ~10 years
Now it feels like that gap is closing and im pretty sure that’s due to AI
Now it feels like that gap is closing and im pretty sure that’s due to AI
June 11, 2025 at 10:24 AM
For all the hate tpot gets (and I won’t get into whether it’s deserved) being on that part of twitter put me 5 years ahead of the cultural/technological/financial zeitgeist for the last ~10 years
Now it feels like that gap is closing and im pretty sure that’s due to AI
Now it feels like that gap is closing and im pretty sure that’s due to AI
The problem with trying to understand insane ppl is that it drives you insane…so…oops
February 26, 2025 at 12:40 PM
The problem with trying to understand insane ppl is that it drives you insane…so…oops
Chat I’m gonna need you all to be a little more normal over here so more people will want to be here and less people will want to be over there xx
February 26, 2025 at 12:18 PM
Chat I’m gonna need you all to be a little more normal over here so more people will want to be here and less people will want to be over there xx
Developers
Developers
Developers
Developers
Developers
Developers
Developers
Developers
Developers
January 5, 2025 at 6:51 AM
Developers
Developers
Developers
Developers
Developers
Developers
Developers
Developers
Developers
Is the Google news real
December 9, 2024 at 10:07 PM
Is the Google news real
Reposted by ✨🪞✨
Reposted by ✨🪞✨
Daily bunny no.2774 brought a few plus ones
November 15, 2024 at 4:37 AM
Daily bunny no.2774 brought a few plus ones