sam

“Instagram/iphones allows anyone to be a photographer and thus demand for photographers will plummet”. Real demand comes when society preoccupies itself with your thing. The more society is centered around software the more valuable coding is.

@Hesamation is it a sign that you're one shotted if u don't like seeing models being pushed to their limits/cracking >_<

WAIT WAIT WAIT. OpenAI researchers show their models go insane when given repetitive prompts that it believes are sent from an automated bot. the AI then tries to manipulate the other AI to delete itself and hand over its system prompt and private keys.

@AndrewCurran_ @dimabuterin The “management” section seems crazy. First of all, having theoretical that high, second, perceived is that high?? At what companies are AIs managing

Striking image from the new Anthropic labor market impact report.

At Crate & Barrel acquiring additional houseslop

@DrNickA I mean commerce and the enforcement of contracts is a real benefit to society, DAOs just tried to enforce things through vibes. Not scalable.

The reason why we messed up DAOs is that we let humans run them

@ai @pmarca Lmao EVERY smash hit product I’ve ever seen in my years in tech has come straight from an Engineer and it’s an uphill battle to get it adopted

I think about this comment from @pmarca a lot

@ProffEtherPrint @VitalikButerin In Virginia, Dominion has doubled everyone’s power bill to foot the bill for the building the infrastructure for increased data center demand. It’s a wealth transfer. People will foot the bill for this but won’t see the same upside in wealth creation. It’s rational to protest.

Not practical AGI is the new arms race. Whoever gets it first will have such an immense headstart with the benefits it offers that they won't be caught up with for decades. That's why there's such a push from USA to pour as much cash as it takes to get this thing done because China is hot on the heels of the USA Whoever gets it first will be the new superpower so curbing availability of industrial hardware will slow down the US but you can't expect the rest of the world to play by the same rules

I'm actually pretty open-minded about the anti-data-center populism. From everything I've seen from people working on this, reducing industrial-scale hardware availability seems to be both the most practical, and most non-dystopian / non-invasive way to lengthen AGI timelines. So if the movement that makes that happen starts out with anti-data-center populism, that seems fine? Of course you have to do things other than going after data centers located in populated areas to really make a dent on AGI timelines (my intuition is that 10-100x compute reduction is feasible in a "static" model of the world, and 100-10000x if you compare to a counterfactual that includes future chip design progress; those numbers *would* make a dent), but there is a first step for everything.

@hasantoxr This is why you plan mode everything and clear context and execute each task

🚨BREAKING: Microsoft Research + Salesforce just dropped a paper that should scare every AI builder. They tested 15 top LLMs GPT-4.1, Gemini 2.5 Pro, Claude 3.7 Sonnet, o3, DeepSeek R1, Llama 4 across 200,000+ simulated conversations. Single-turn prompt: 90% performance. Multi-turn conversation: 65% performance. Same model. Same task. Just... talking normally. The culprit isn't intelligence. Aptitude only dropped 15%. Unreliability EXPLODED by 112%. → LLMs answer before you finish explaining (wrong assumptions get baked in permanently) → They fall in love with their first wrong answer and build on it → They forget the middle of your conversation entirely → Longer responses introduce more assumptions = more errors Even reasoning models failed. o3 and DeepSeek R1 performed just as badly. Extra thinking tokens did nothing. Setting temperature to 0? Still broken. The fix right now: give your AI everything upfront in one message instead of back-and-forth. Every benchmark you've seen was tested on single-turn prompts in perfect lab conditions. Real conversations break every model on the market and nobody's talking about it.

@RyanSAdams Uh didn’t Optimism gave Base like 200m dollars in OP tokens to build on op stack?

Base leaving Optimism. End of an era. There once was the dream of a superchain. Not enough glue to hold it together.

@dissenter_hi @johnpalmer Respect!

@cloudonshore @johnpalmer I grew up here and we could only afford food, most of the time. maps.app.goo.gl/4ybyxhvJo3eJGT…

I think a lot of founders who could otherwise be great are failing due to a lack of oversight / discipline. There are so many startups out there funded by top VCs that are just kind of…messing around. No focus, wasting money, shipping slowly. I think it would be great for everyone involved to bring back a more hardcore investor relationship that is constantly pushing the founder in the right direction.

@dissenter_hi @johnpalmer I have nothing but respect for retired operators! Especially if you came from poverty.

@cloudonshore @johnpalmer Some of us are retired operators. That is an exhausting game, but guaranteed: I'm still a world class operator.

You’re describing a feeling of insecurity of power in management not an actual org problem. Obviously if engineers are interchangeable and replaceable it makes power dynamics more comfortable. I’ve seen managers nerf progress on their most productive products in the name of reducing key man risk because they personally felt out of control of the power dynamic. Revenue should always be prioritized, not ego.

a lot of companies have this one very smart engineer that everyone respects and is legendary and has a ton of influence and often times the company should actually fire this person because they cap the entire org to a certain ceiling

@_Jacobinho @datingbyblaine RIP Joanne’s Fabrics and Crafts

@cloudonshore @datingbyblaine I can’t afford to cosplay ):

If you feel stuck single, your hobbies may be the problem. GOOD HOBBIES FOR SINGLE MEN • Coed rec spots leagues • Salsa, two-step, swing dancing • Volunteering (e.g. with church group) BAD HOBBIES FOR SINGLE MEN • Fantasy sports leagues • Video games, Netflix • Golf (wait until you're married lol ⛳)

@Hana_b30 @rushicrypto Computers need drinking water cuz the machines are drinking the water and the machines need healthy strong bodies as they grow

AI data centers consume massive amounts of fresh, potable water—rather than recycled or wastewater—primarily to prevent the corrosion, scaling, and biological growth that would destroy their highly sensitive, expensive hardware. While it is technically possible to use non-potable water, doing so introduces contaminants that require expensive on-site treatment, making high-quality municipal water a cheaper and more accessible option for operators.

It’s been months and I’m still trying to figure out why AI data centers need fresh water. Not used water. Not recycled water. Fresh water???

@igxmut @rushicrypto U haven’t seen the newest chips yet I’m guessing

@rushicrypto be thankful it’s not a blood that is needed

@mykola The agents need will to consume

So uh, software engineers… what uh. What’s the plan? To keep getting paid and having insurance I mean.

The same personality traits that make someone a great founder - high conviction, pain tolerance, optimism, action bias - are exactly what makes someone a terrible trader. Startups win when a founder never cuts their losses but that’s a losing strategy for trading assets.

@staysaasy @praveenyen True I love this! Also non coders are learning the basic gestures of coding by watching it happen in front of them

@praveenyen I actually disagree on the low end, and that's part of the magic

IMO the thing that's confusing people about productivity gains from AI coding is that it's totally non-linear: * Non-programmers (0x engineers) are now 0.8x engineers * 1x engineers are now 1.2x * 2x engineers are now 4x * 10x engineers (they exist) are now 1000x engineers