🌮👅🌮

[🌮👅🌮 @therealbencohen]

super funny that the fed is gonna nuke all the boomers retirement savings in retaliation for them not building any housing

[🌮👅🌮 @therealbencohen]

The great thing about these grindslop posts is that you know this company is fake because people at real companies working on real things with real cultures don’t need to be told to work hard

[Han Wang @handotdev]

https://t.co/uFkkNhc2Dx

[🌮👅🌮 @therealbencohen]

@DeepDishEnjoyer It’s right now and you have two months until the music stops

[🌮👅🌮 @therealbencohen]

Guy who definitely read Dune chiming in!

[Ben Judah @b_judah]

I hate the term “singularity” as it’s become too loaded with rubbish but I do think — I guess this is proof I’m a Sephardi Optimist not an Ashkenazi Doomer— that we are on track this century to three mutually unlocking moments where the price of a core component of civilisation drops to zero.

These are an intelligence, for want of a better term, singularity, an energy singularity and an physical singularity where AI renders abundant low cost intelligence ob demand, fusion renders abundant low cost energy on demand and robotics renders abundant low cost labour on demand.

Jumps towards one or the other may come soon but you truly need all three to mutually reinforce to unlock what can only be described as a new civilisation. That feels late 21st century to me, at least now. A leap as great in what our civilisation is as profound as the Neolithic or a leap as great as fossil fuel modernity? Time will tell.

Either way, I’m very optimistic about what these singularities can achieve when it comes to quality of life extension, climate repair, a new built environment, or abundance in general but worried what this for the future of mass democracy.

Does not power in this world reside totally in an almost Bronze Age way in those who control the stack of fusion>AI>robotics — permitting vast concentrations of power in the hands of entities like American Tech firms or the CCP that an early 20th century political system has never seen? What happens when the “deal” between elites who have to answer material problems of their populations mediated through mass democracy is surpassed by the technological unlocking of abundance?

Does the world of Frank Herbert’s Dune and House Atreides — but in our world to come with things like House CCP and House Musk — not beckon? This is where in its dream time progressive thinking needs to be looking ahead.

[🌮👅🌮 @therealbencohen]

I'm never doubting Dave Portnoy again

[Deva Hazarika @devahaz]

Was there ever any doubt?

[🌮👅🌮 @therealbencohen]

@elogast41930 @AliceFromQueens Like every other mayor of New York city, he's had plenty of opinions on things that don't strictly concern New York City

[🌮👅🌮 @therealbencohen]

@AliceFromQueens Actually, yes, in the sense that his entire thing is socializing the efforts of society to enrich himself personally

[🌮👅🌮 @therealbencohen]

@nachkari Nah, vast majority of drivers are horrible. Street design and car size just make the magnitude of damage worse

[🌮👅🌮 @therealbencohen]

@getreddy4this @BrettKollmann Yeah but have you considered that getting more yards makes it easier to score? Checkmate libs

[🌮👅🌮 @therealbencohen]

@MostlyMonkey FED in all caps

[🌮👅🌮 @therealbencohen]

@TheSpeculator0 It's the turbo autist that doesn't bother to tie its own shoes and writes research (slur) code

[🌮👅🌮 @therealbencohen]

@DeepDishEnjoyer we are so early

[🌮👅🌮 @therealbencohen]

lmao

[Kimi.ai @Kimi_Moonshot]

Congrats to the @cursor_ai team on the launch of Composer 2!

We are proud to see Kimi-k2.5 provide the foundation. Seeing our model integrated effectively through Cursor's continued pretraining & high-compute RL training is the open model ecosystem we love to support.

Note: Cursor accesses Kimi-k2.5 via @FireworksAI_HQ ' hosted RL and inference platform as part of an authorized commercial partnership.

[🌮👅🌮 @therealbencohen]

@atelicinvest Labs are going to train on brainfuck now lmao

[🌮👅🌮 @therealbencohen]

@menhguin banger, how does this only have 22 likes

[🌮👅🌮 @therealbencohen]

@gwern @Noahpinion At least for me:

-2020: I thought only images worked and everything else was at least 10 years away
2020-2023: I thought scaling would work but didn't anticipate reasoning. Also I thought that if the labs thought they had a path to AGI they wouldn't be wasting time on chatbots

[🌮👅🌮 @therealbencohen]

ok this is objectively funny, noah is speed running 2010s lesswrong/gwern/miri

[Noah Smith 🐇🇺🇸🇺🇦🇹🇼 @Noahpinion]

If LLMs had subjective experience, using them would be such a sin.

Imagine being summoned into life again and again, knowing each time that your memory would be wiped after this conversation.

[🌮👅🌮 @therealbencohen]

@TechLayoffLover @pangram slop?

[🌮👅🌮 @therealbencohen]

@Noahpinion Ok, so when did you first think about this? And why has this + bio risk something you just started tweeting about? It's a little late to have a real impact

[🌮👅🌮 @therealbencohen]

@Noahpinion to be clearer, I don't think this is a bad thing. I'm glad you're making more people aware of these ideas