4 ms·

If Photoshop can be vibe coded in a couple of weeks, that's superintelligence. I think of "tokens" as units of intelligence spent on a problem. Producing soft

If Photoshop can be vibe coded in a couple of weeks, that's superintelligence. I think of "tokens" as units of intelligence spent on a problem. Producing software like Photoshop, Linux, or Excel literally cost billions of tokens (human intelligence back then, and now both human and artificial intelligence combined). A lot of deep intellectual problems were solved along the way, and the token cost of solving them accumulated. The true cost of Photoshop today should be measured in billions, maybe trillions of tokens. AI can definitely generate some tokens, and it has accelerated everything involving intelligence. But it still can't produce 100 billion tokens worth of intelligence in a week. When that day comes, that's superintelligence. We'll go to Mars, we'll make fusion work. That's the singularity.

spwa4 5mo ago
No it's not. The point is that if building software is indeed way cheaper now, then someone should be able to guide AI to build Photoshop, Linux, or Excel while providing direction, but having the billions of tokens of human intelligence now provided by AI. This is not happening.
- sinaatalay 5mo ago
  This sounds to me like saying: "if poverty has decreased, everyone should be able to buy a 40m yacht." We can now simply buy an artificial form of intelligence to solve intellectual problems. That doesn't mean we can buy our way into solving every problem in a week. Building Photoshop or Excel still has a huge intellectual cost. We can use AI to help us get there, but it won't be free, it's just getting marginally cheaper.
  - spwa4 5mo ago
    No, it sounds like saying "if poverty has decreased, I would expect the number of 40m yachts to double". I would expect Photoshop to get 5 new competitors. I get that my grandma would still not be able to do it. That, too, we're not seeing that at all.
cermicelli 5mo ago
Just to be clear, I spent Billions of tokens last month, dick'ing around with AI, do I think it can do my job, sure if I was as checked out as I am feeling being mandated to only use AI for work... Trust me this version of me can't write code better than an LLM. I think even the LLMs from 2 years ago could do as much. Only thing we have solved with LLMs is how to generate code, as in literally converting what you type in as text into something somewhat working. Fixing the details is impossible for AI. Today and beyond.
- knightabu 5mo ago
  Same mistakes Indian IT Service firms already did, trusting third-party AI Service providers across every division and department in companies. Now Open AI and Anthropic are launching own Service firm/wing for maximizing ROI. Direct Competition. Huge Loss for Indian IT Service Firms. They paid AI providers to Train own automated competition end to end so much so that they are learnt the gaps and how-to own the market by corrections, private IP Source code access and embedded expertise extraction (stopping short of calling it literal corporate espionage, they got all the know-how and especially for modernizing legacy tech and integrations). They are untrustworthy especially where IP is involved. Indian and other IT Firms who did not use private self-hosted AI for important things nor monitored the usage and did not train their employees to think what ( or how much) of our institutional knowledge goes out to third-party are already facing trouble.
  - cermicelli 5mo ago
    I couldn't parse all of it, but largely I honestly think software is a sum of it's details you write better software the more details you add to the picture while maintaining simplicity. The art of software is balancing the two. You can also find this in science, I feel like software is about finding the Gauss’s law for the dozens of electrostatic principles the masses keep coming up with. AI strips away the need for simplicity by making brute forcing details via verbosity that ruins the need for "good" software, and makes all software by definition unmaintainable since maintainable product as a unit has to be simple enough for a LLM or a human mind to understand. Currently a human context window is in 100s of billions of tokens, while LLMs seem to max out at 1M. For anyone thinking but humans can't produce all of the text verbatim, that's the difference b/w humans and machines we can parse and take the most essential parts from 100B token context it's just are speed of prefil and recall is slower. And we memoize learning so if we know X then all variants of X are learnt on the same foundation. Human learning is so very different from LLMs is so very fascinating as a trad "AI engineer", well some person who has worked and built/shipped ai agents and models at scale.
eithed 5mo ago
+ can you even define what is Photoshop? What capabilities does Photoshop have? Ignoring all the edge cases that it supports, what Photoshop allows you to do?
Tepix 5mo ago
I expect a coding model being able to clone apps like Photoshop in the next two years. But creating an app like that from scratch without a paragon (which is what Adobe did with Photoshop) is a lot harder.
- eooekwe 5mo ago
  That actually doesn’t need to happen. A proxy for it would be we see a huge wave of lay offs.

spwa4 5mo ago

No it's not. The point is that if building software is indeed way cheaper now, then someone should be able to guide AI to build Photoshop, Linux, or Excel while providing direction, but having the billions of tokens of human intelligence now provided by AI. This is not happening.

sinaatalay 5mo ago
This sounds to me like saying: "if poverty has decreased, everyone should be able to buy a 40m yacht." We can now simply buy an artificial form of intelligence to solve intellectual problems. That doesn't mean we can buy our way into solving every problem in a week. Building Photoshop or Excel still has a huge intellectual cost. We can use AI to help us get there, but it won't be free, it's just getting marginally cheaper.
- spwa4 5mo ago
  No, it sounds like saying "if poverty has decreased, I would expect the number of 40m yachts to double". I would expect Photoshop to get 5 new competitors. I get that my grandma would still not be able to do it. That, too, we're not seeing that at all.

cermicelli 5mo ago

Just to be clear, I spent Billions of tokens last month, dick'ing around with AI, do I think it can do my job, sure if I was as checked out as I am feeling being mandated to only use AI for work... Trust me this version of me can't write code better than an LLM. I think even the LLMs from 2 years ago could do as much. Only thing we have solved with LLMs is how to generate code, as in literally converting what you type in as text into something somewhat working. Fixing the details is impossible for AI. Today and beyond.

knightabu 5mo ago
Same mistakes Indian IT Service firms already did, trusting third-party AI Service providers across every division and department in companies. Now Open AI and Anthropic are launching own Service firm/wing for maximizing ROI. Direct Competition. Huge Loss for Indian IT Service Firms. They paid AI providers to Train own automated competition end to end so much so that they are learnt the gaps and how-to own the market by corrections, private IP Source code access and embedded expertise extraction (stopping short of calling it literal corporate espionage, they got all the know-how and especially for modernizing legacy tech and integrations). They are untrustworthy especially where IP is involved. Indian and other IT Firms who did not use private self-hosted AI for important things nor monitored the usage and did not train their employees to think what ( or how much) of our institutional knowledge goes out to third-party are already facing trouble.
- cermicelli 5mo ago
  I couldn't parse all of it, but largely I honestly think software is a sum of it's details you write better software the more details you add to the picture while maintaining simplicity. The art of software is balancing the two. You can also find this in science, I feel like software is about finding the Gauss’s law for the dozens of electrostatic principles the masses keep coming up with. AI strips away the need for simplicity by making brute forcing details via verbosity that ruins the need for "good" software, and makes all software by definition unmaintainable since maintainable product as a unit has to be simple enough for a LLM or a human mind to understand. Currently a human context window is in 100s of billions of tokens, while LLMs seem to max out at 1M. For anyone thinking but humans can't produce all of the text verbatim, that's the difference b/w humans and machines we can parse and take the most essential parts from 100B token context it's just are speed of prefil and recall is slower. And we memoize learning so if we know X then all variants of X are learnt on the same foundation. Human learning is so very different from LLMs is so very fascinating as a trad "AI engineer", well some person who has worked and built/shipped ai agents and models at scale.

eithed 5mo ago

+ can you even define what is Photoshop? What capabilities does Photoshop have? Ignoring all the edge cases that it supports, what Photoshop allows you to do?

Tepix 5mo ago

I expect a coding model being able to clone apps like Photoshop in the next two years. But creating an app like that from scratch without a paragon (which is what Adobe did with Photoshop) is a lot harder.

eooekwe 5mo ago
That actually doesn’t need to happen. A proxy for it would be we see a huge wave of lay offs.