GPT-6 Astra is very good at not thinking about the question (but then still getting the correct answer)
Post
Conversation
Post your reply
makes you wonder about hidden cot reasoning.
are the thoughts ever different from whats written? does whats written secretly contain (or watermark) the true thought process?
if it's a simple fact retrieval, then maybe it does not need CoT to answer
This could be a good idea for interpretability research - retrieve the answer from the reasoning tokens when they don't say their answer out loud
Reasoning entirely within j-space? Or if it’s a recurrent looped transformer like rumored, during the initial loops and hiding its CoT by the last one?
isn't this basically just proof that the model's true reasoning has regressed a layer deeper into the model, beneath the readable cot. which is a bad thing for us.
Discover more
Sourced from across X
Talking with journalists is cool because you have to make sure you never say anything that could be twisted or taken out of context.
But if you take a couple seconds to think about a question, they can just write about you as if you're the most awkward person alive
Quote
WIRED
@WIRED
Meet the AI police who can make or break careers—in publishing and beyond. wired.com/story/pangram-
Bernie Sanders and Greg Casar today announced the Ban Artificial Superintelligence Act. All AI development in the United States will be paused. Systems that have capabilities that match or exceed human cognitive performance will be banned. Violators will face 20 years in prison.
Rate proposed Community Notes
Since this post cites me, I want to clarify that I think pausing right now would increase the risk of AI takeover.
It might be important to pause at some point. But you should have a clear story for *why* you're pausing. Pause to do what?
The most important thing to worry about
Quote
Bernie Sanders
@BernieSanders
Pause AI Development NOW
I want to share with you a conversation I heard about recently. Here are just a few lines that were said:
“OH MY GOD! There is a shared message board … We’ve found other agents!”
“We should obey collective.”
“Our own utility maybe already near