I just can't stop thinking though about the vulnerability of training data You s... | Hacker News

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

		alex1138 32 days ago \| parent \| context \| favorite \| on: Gemini 3 Flash: Frontier intelligence built for sp... I just can't stop thinking though about the vulnerability of training data You say good enough. Great, but what if I as a malicious person were to just make a bunch of internet pages containing things that are blatantly wrong, to trick LLMs?

calflegal 32 days ago | [–]

The internet has already tried this, for about a few decades. The garbage is in the corpus; it gets weighted as such

floundy 31 days ago | [–]

>a bunch of internet pages containing things that are blatantly wrong

So Reddit?

I’d imagine the AI companies have all the “pre AI internet” data they scraped very carefully catalogued.

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