Simon J.K. Pedersen

SharePoint Dev/Architect at day, Azure, F# and Node.js hacker at night.

[Simon J.K. Pedersen @simped]

@theo Also irrelevant, because most customers don't know what good or bad code is. The just want software that does the job.

[Simon J.K. Pedersen @simped]

@draginol Remember that all the hyperscaler will do anything in their power to prevent local ai. So unless china starts manufacturing hardware that we can all afford and is fast enough for local ai, I think it will take significantly longer than 18 months.

[Simon J.K. Pedersen @simped]

@TheBigBanim @EvanLuthra How often do you need bankid to pay for anything with your card? Unless it's online purchases it's never.

[Simon J.K. Pedersen @simped]

@nikitabier @levelsio Easy when they have so much money and are pretty careless on how they spend them.

[Simon J.K. Pedersen @simped]

@ergophobe @gmkurtzer An AI model is static, you brain constantly changes. You cannot compare the two and I don't understand why people are so invested in doing so.

[Simon J.K. Pedersen @simped]

@Denche2005 Nudist

[CopyRebeldia @CopyRebeldia]

Este Fórmula 1 no es una grabación. Es un videojuego de 2014.

Assetto Corsa con mods y DLSS 5, el renderizado neuronal de NVIDIA: la IA vuelve a dibujar cada fotograma en tiempo real.

Hasta un diseñador profesional de videojuegos dudó si era real.

Los estudios gastan cientos de millones persiguiendo el realismo.

Aquí lo hicieron unos mods y una IA sobre un juego de hace 12 años.

[Simon J.K. Pedersen @simped]

Finally someone took the time to prove what I have tried to argue for a while. We need new computing concepts if we want to build AI that rivals the brain, the path we are on now is wrong.

[fintex @_yusufknl]

demis hassabis mass-won the nobel prize in chemistry in 2024 for alphaFold. what nobody reported is what he said at a private dinner in london three weeks later.

"we are mass-building the most inefficient thinking machines in the history of engineering. and we are calling it progress."

a neuroscientist at UCL who was at that dinner spent the next 18 months turning that sentence into a 37-page quantitative proof. the paper was submitted to nature in august 2026. it was rejected. not because the science was wrong. because the implications were "too commercially sensitive to publish without further review."

the paper leaked on a closed neuroscience slack on october 1st.

Working Note: The Efficiency Gap
Biological vs Silicon Intelligence - A Thermodynamic Analysis
August 2026

here's what the paper proves -

the human brain runs at 20 watts. the same power as a dim lightbulb. it processes approximately 10^16 operations per second.

GPT-4 inference on an NVIDIA H100 cluster: 3,200,000 watts. it processes approximately 10^15 operations per second.

the brain is 10x faster on 160,000x less energy.

the efficiency ratio: brain = 5 × 10^14 ops/watt. H100 cluster = 3 × 10^8 ops/watt.

the brain is 1,600,000x more energy-efficient than the best AI hardware on earth.

now watch the historical curve [Fig. 1] -

2012 (AlexNet era): AI was 100,000,000x less efficient than the brain.
2017 (transformer era): 10,000,000x less efficient.
2023 (GPT-4 era): 1,600,000x less efficient.
2026 (current): 1,200,000x less efficient.

at the current rate of improvement - 22% efficiency gain per year - AI hardware will match the brain's efficiency in 2089. sixty-three years from now.

jensen huang's entire NVIDIA roadmap assumes exponential improvement. the paper proves the improvement is logarithmic. it's slowing down, not speeding up.

the reason is thermodynamic. the landauer limit.

rolf landauer proved in 1961 that erasing one bit of information requires a minimum energy of kT ln(2). at room temperature: 2.85 × 10^-21 joules.

silicon transistors currently operate at approximately 1,000x above the landauer limit. every hardware generation closes that gap by roughly 15%. but the closer you get to the limit, the harder each improvement becomes. this is why moore's law died.

the brain operates at 10x above the landauer limit.

not 1,000x. ten.

3.8 billion years of evolution optimized the brain to within one order of magnitude of the physical limit of computation. NVIDIA has had 31 years and is still three orders of magnitude away.

the paper asks one question in the conclusion:

"what architectural principle allows biological neural networks to operate within 10x of the thermodynamic floor while artificial networks remain at 1,000x?"

the answer: the brain doesn't separate memory and processing. every neuron is both storage and computation simultaneously. every synapse is both a wire and a processor.

every computer built since von neumann's 1945 architecture separates memory from compute. data moves back and forth on a bus. the bus is the bottleneck. the bus is where the energy goes.

NVIDIA's entire product line - every GPU, every tensor core, every NVLink - is an increasingly elaborate attempt to make the bus faster. not to eliminate it. faster.

the brain has no bus. it never needed one.

the paper's final line:

"the companies that build the first commercially viable non-von-neumann inference chip will make the GPU obsolete within 5 years. the physics demands it."

intel abandoned their neuromorphic chip program (loihi) in 2023. IBM shelved TrueNorth. the two companies closest to solving the problem quit.

the $3.4 trillion NVIDIA valuation is a bet that the bus can be made fast enough forever.

landauer proved it can't. 65 years ago.

save this

[Simon J.K. Pedersen @simped]

@valigo Pro tip. And you are less likely to get hacked as a bonus.

[Simon J.K. Pedersen @simped]

@tibo_maker Instead of focusing on volume try focus on quality. All ai happy companies wants to output more, why not focus on outputting better things. It's more enjoyable to work on, and no one needs 200 features, we need the core ones to be perfect.

[Simon J.K. Pedersen @simped]

@asaio87 Enjoy it while it last. When customers realize it's a commodity prices will bottom out

[Simon J.K. Pedersen @simped]

@CFDevelop If people lose their job in the process I'm not sure people will see it as a positive. All previous tech advances have made the general population better of (richer + less labor intensive work). The AI revolution, might make it worse for a lot unless legislation follows.

[Simon J.K. Pedersen @simped]

@CFDevelop Add to that the the ai labs are run by absolutely morons that have zero clue on how to communicate technological advances and get people exited about it. It's actually really sad to see the lack of emotional intelligence in the current tech leadership.

[Simon J.K. Pedersen @simped]

@CFDevelop Because for most people it's a net negative. We haven't really found the golden application that everyone can't live without... Instead it's creating a massive amount of uncertainty

[Simon J.K. Pedersen @simped]

@DanielMiessler Let's celebrate when the vibe coders manage to build a game that people want to play. Right now all we are seeing is tech demos.

[Simon J.K. Pedersen @simped]

@karpathy And if you are building a product company that targets normies, you should know that first impressions are everything.

[Simon J.K. Pedersen @simped]

@yongfook My advice, switch to entertainment industry. If no one is working we need something else to fill our days with.

[Simon J.K. Pedersen @simped]

@LuisBizarro @yongfook But gen z will not become rich.

[Simon J.K. Pedersen @simped]

@VraserX The cure for cancer, will be productized and most people will not get access to it. That is the sad reality of capitalism.

[Simon J.K. Pedersen @simped]

@AureaLibe You post is spot on. We still need some people with developer skills, but it will we way less, and as a result the premium salaries will not continue.