\
  The most prestigious law school admissions discussion board in the world.
BackRefresh Options Airdrop

How long until China totally leapfrogs the USA in AI? < 1 year cr?

...
boyish lime kitchen digit ratio
  07/01/26
It probably already has Reminder most common "Americ...
Fragrant drunken faggot firefighter brunch
  07/01/26
If it has then where is the model?
wonderful khaki depressive
  07/01/26
In ur penis probably
Fragrant drunken faggot firefighter brunch
  07/01/26
distillation is overblown but chinese models still benefit f...
massive razzmatazz property faggotry
  07/01/26
seems reasonable
Excitant insanely creepy stage
  07/01/26
Distillation is overblown if being treated as a full magic c...
wonderful khaki depressive
  07/01/26


Poast new message in this thread



Reply

Date: July 1st, 2026 4:38 PM
Author: boyish lime kitchen digit ratio



(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974123)



Reply

Date: July 1st, 2026 4:41 PM
Author: Fragrant drunken faggot firefighter brunch

It probably already has

Reminder most common "American AI" workers name is Ching Chong

(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974133)



Reply

Date: July 1st, 2026 4:42 PM
Author: wonderful khaki depressive

If it has then where is the model?

(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974136)



Reply

Date: July 1st, 2026 5:46 PM
Author: Fragrant drunken faggot firefighter brunch

In ur penis probably

(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974295)



Reply

Date: July 1st, 2026 4:51 PM
Author: massive razzmatazz property faggotry

distillation is overblown but chinese models still benefit from being able to distill and train on synthetic data produced by US frontier models. i feel like if it were easy for them to completely catch up they would have already given how many companies are attempting to do so there

the more relevant question is "when do chinese models get good enough that the difference doesn't practically matter for 99% of use", which i think we're 1-2 years away from. sooner for coding specifically.

(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974150)



Reply

Date: July 1st, 2026 6:09 PM
Author: Excitant insanely creepy stage

seems reasonable

(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974325)



Reply

Date: July 1st, 2026 6:10 PM
Author: wonderful khaki depressive

Distillation is overblown if being treated as a full magic catch-up. But underblown as an explanation for rapid practical convergence. And it is definitely cheating big time, although probably not legally exact to call it "counterfeiting" it is more in that direction than anything else.

Also I think the part you are missing is that a model has not "caught up" just because it meets benchmarks if it can only generate better capability through privileged access to a stronger teacher. That just makes it a successful parasite.

(https://xoxohth.com/thread.php?thread_id=5879021&forum_id=2#49974332)