Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> Thanks to our advanced AI algorithms, you won’t tell generated humans from real people

If the images posted are the best they can do, then i have some bad news from them



I can count on one hand the number of ways these photos fail. That's right, 6 ways.


If you encode a binary digit for each biological digit on your hand you can count up to 32 on one hand.


and there are more than two clearly distinct ways to articulate a finger, so you can bump that up to at least 3^5, not that it would be in any way convenient or useful to do so

assuming you have the dexterity to hold each of your 10 fingers in one of 3 different positions—e.g.: up, down and crooked—you could technically count up to 3^10 or 59,049

if you mastered that, you could add a fourth or even fifth position. 4^10 is just over 10^6, so you could hold any 6 digits, and 5^10 is just under 10^7, so almost any 7 digits

you can compress really heavily using high bases, but for it to make practical sense the atomic medium of storage has to have that number of states

if you just convert binary to a high base and back again, unsurprisingly the space used works out exactly the same, as the number of symbols used to represent a number may be fewer, but the length of the symbols themselves balances it out exactly

e.g. in base 1000 you may be able to represent the number 999 with a single symbol, but you need 1000 unique symbols to show every number up to 999, each of which of course takes up significantly more digits than a bit; however, if you had a storage cell that could reliably store and display 1000 unique states, then it would be a different matter. whether that's possible or would/could even take up less physical space than ~log_2(1000) binary cells I do not know


Nono, these AI improvements mean that we can do 64 now!


Only 31 right? 16 + 8 + 4 + 2 + 1


up to 31, but 32 numbers in total


Count Rugen sees no problem with this.


Eh. On one hand, I guess it's no problem at all. But on the other hand...


Marketing taking it too far as usual. They certainly have mastered peak uncanny valley though, I'm not really sure what this is useful for.


The first photo generated for me made everything look plastic. Unnatural sharp lines on everything. Shadows from 5 different directions.

It’s laughable to call these “hyperrealistic”.


The first image i generated was worse than that

A mermaid with plastic looking skin, and badly rendered ocean water in background.

https://generated.photos/human-generator/64d6dde03af7f90007c...


I got exactly that image too! I guess the "random human" isn't so random. This calls their "real time" claim into question...


Looks as realistic as every other real mermaid I’ve seen ;-)


It nailed the pose, though.


The demo pictures have this sort of "uncanny valley" effect for me. I wonder if it happens when I don't know that the pictures are AI-generated.


In the second photo I rolled, the generated human had legs that ended in hands instead of feet.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: