(developer.apple.com)

485 points zora_goron | 1 comments | 29 Aug 25 00:44 UTC | HN request time: 0.342s | source

Show context

echelon ◴[29 Aug 25 01:31 UTC] No.45059007[source]▶

Can't each of these companies with IDE integrations slurp up the network traffic and distill Anthropic's models?

If you can listen to billions of tokens a day, you can basically capture all the magic.

adastra22 ◴[29 Aug 25 07:24 UTC] No.45061207[source]▶

That is not how training works…

replies(1): >>45063596 #

1. echelon ◴[29 Aug 25 13:05 UTC] No.45063596[source]▶

That's how model distillation works.

DeepSeek is the most notable case, but it's been used lots.

And the foundation model companies are scraping and exfiltrating each others' data.

Claude Sonnet will ship in Xcode