Reddit Wants Bing, Anthropic, and Perplexity to Pay To Search the Site

[engine]

Reddit has struck a deal with google and OpenAI, the CEO, Steve Huffman, and wants others to pay to continue to scrape Reddit.

�Without these agreements, we don�t have any say or knowledge of how our data is displayed and what it�s used for, which has put us in a position now of blocking folks who haven�t been willing to come to terms with how we�d like our data to be used or not used,� Huffman said in an interview this week. He specifically named Microsoft, Anthropic, and Perplexity for refusing to negotiate, saying it has been �a real pain in the ass to block these companies.�

[theverge.com...]

www.theverge.com Reddit CEO says Microsoft needs to pay to search the site Reddit says Microsoft's Bing, Anthropic, and Perplexity have scraped its data without permission. -�It has been a real pain in the ass to block these companies.-�

[londrum]

i wonder how long search engines will keep training their AI on reddit.

they're probably just using it for language learning, rather than up-to-date news. surely once they've been through it once then they'll have diminishing returns.

it will be interesting to see if there comes a time when google decide to stop paying reddit millions, and then there'll be no reason for them to artificially inflate reddit's serp's position. their traffic could plummet as quickly as it rose

[engine]

I suspect you're correct: once trained, is there any need to keep training.
with news, however, that's different.

[doc_z]

Indeed, it's "a real pain in the ass to block these companies". It's like a race between the hare and the hedgehog. In my opinion, the better way is to throw sand in the AIs' gears and set traps...

When he talks about "our data", he is probably technically and legally right. From my point of view, however, the data belongs to the community.

[sudo]

Among other reasons, Reddit is uniquely great for training because of how it's structured. The reason they are charging is likely because of the technical burden...imagine trying to keep one of the largest sites on the internet up while constantly getting pumped for training data.