The second chapter of Situational Awareness, can be summarized in the following sentence: 👉 The A-bomb is to H-Bomb what AGI is to ASI. As Leopold says, “The Bomb was a more efficient bombing campaign. The Super was a country-annihilating device. So it will be with AGI and Superintelligence.” Aschenbrenner makes a good analogy at the beginning of the chapter, comparing the evolution from Atomic bombs to hydrogen bombs, which would be directly proportional to what AGI is to ASI. A great point was made: when the H-bomb was developed - a weapon vastly more destructive than all WWII bombs combined - nuclear policies and war plans stayed stuck at the A-bomb level, completely missing the exponential leap in power and impact.

@AmbsJam Is it a middle-ground? I liked what I heard from Sinclair and then Attia was completely against everything Sinclair said. So I was left …

RT @CreatedByJannn: @BenjaminDEKR you literally ranked Grok 3 alongside competitors and it is not even released. Huge red flag for an inter…

RT @WhoTFIsHatman: Hey @HBO, I have an idea.

@AISafetyMemes Its a AI-bro code.

I’ve said this as a client before 😂 but right before moving on. If the design is bad because of taste preferences, fine, we can work on it. But if the basics are wrong (bad alignment, inconsistent spacing, zero UX sense), no amount of revisions will fix it. Just wasting time at that point. So if that’s the best you can offer, thanks, we love it, no change needed from your side.

@bsngln38133 Why Claude is not on the list!? I feel offended.

ChatGPT had an identity crisis thinking IT WAS CLAUDE and totally gaslit me! WTF!

Can we stop using purple? Every time I see it, I already know … it’s some SaaS that will waste my time and money. Clean UI, terrible UX. Lately, every AI tool or SaaS platform looks the same. ‘We do this, we do that’ - but in the end, they’re mostly useless, overhyped, and frustrating to use.

@thatguyDOR That’s eyeporn. Very clean and coherent.

@lefthanddraft Haters will always hate.

Claude has different levels of excitement, and they scale directly with the illusion of progress: - Ah, I see the issue now. - AHA! - WAIT! LOOK AT (…) Sometimes he gets way too excited, thinking we cracked the ultimate solution only to backtrack mid-sentence like he is debugging himself in real time. 😂

When I get sick of notion and clickup … Henri can predict our next move. Overcome that.

@RyanEls4

@heysatya_ Left, naturally the eyes scan from left to right, at least in the Western world. The centered one is beautiful, but it’s an overload of information.

@MichaelFilipiuk Elegant!

Why is Hillary Clinton always involved in AI somehow 😂 Attached is a snippet from Karpathy’s video showing “Training language models to follow instructions with human feedback,” where human labelers are guided on what an ideal assistant response should be.

@Abmankendrick A should be illegal.

Maybe if I watch it enough, I’ll learn something.

@lefthanddraft Wow.