Britannica and Merriam-Webster Sue Perplexity Over AI 'Answer Engine' (reuters.com) 20
Perplexity AI is the latest AI startup to be hit with a lawsuit by copyright holders, accused by Encyclopedia Britannica and Merriam-Webster of misusing their content in its "answer engine" for internet searches. From a report: The reference companies alleged in New York federal court on Wednesday that Perplexity unlawfully copied their material and diminished their revenue by redirecting their web traffic to its AI-generated summaries.
Re: (Score:2)
Re:That's cute (Score:5, Insightful)
It doesn't seem to be about the AI being trained on the content, either. It's about the AI searching and summarizing that specific content, without sending any traffic or money their way.
The problem with allowing this is that primary news sources will go broke if only derivative ones get paid.
The problem with not allowing it is an AI that can search and summarize or find answers to specific questions for you is extremely useful.
Re: (Score:2)
Perplexity will set a precedent about agents. We already had Cloudflare not distinguishing between crawlers and agents, now Merriam-Webster.
A crawler goes from page to page, usually saves content to process it later and follows all links that look relevant. It uses robots.txt to judge which links not to follow, just as nofollow attributes and metatags related to that.
An agent is basically a more intelligent proxy for the user's browser. The search engine result is from an index created using the crawler (I
Re: (Score:1)
This is live and continuous, it's not a one-shot affair like torrenting a film and training on that.
"Does fair use include completely MITMing a website's traffic away from them?"
"Should the originator of information have the right to release under a license they choose?"
and a followup relevant to SlashDot, "if you code an
Re: (Score:1)
honestly whatever question i ask an AI is infinitely more important than how much corporations should be getting paid.
burn it all to the ground but until that happens I'm not paying for information and anyone trying to make it harder for information to be freely available is an asshole and people should shit in their hat.
Re: (Score:2)
The follow-up question is what the german publishers are already discussing, if you can force the user to use a certain way to display the content, in that case a browser without content (ad) blockers.
If you rule that the page has to be rendered a reference way (probably like the web designers tested it in Chrome) you open a large can of worms. Do browsers have to run scripts? Do they need to show images? All of these (missing) features would mean rendering the page other than intended, many of them mean no
Re: (Score:2)
Crime pays after all... (Score:4, Insightful)
If you steal someone's property and benefit from it you should be forced to give up every single penny you've earned from it and then you should pay an equal amount in punitive damages.
Otherwise, what good is a 100 million fine if you've earned a billion from the stored content?
The only way to disincentive crime is to stop it from being profitable in the first place.
Move Fast and Steal Shit (Score:1)
nuf sed
Re: (Score:1)
knowledge is not property and i will not be extorted in order to gain access to the wealth of human knowledge.
Re: (Score:2)
Re: (Score:2)
If only (Score:4, Funny)
Facts are not copyrightable (Score:3)
Facts (the substance of encyclopedias) are not copyrightable.
Certain expressions of facts can be copyrightable.
Unless the LLM reproduces the expressions verbatim, they have no case.
Fact of the day: "uncopyrightable" is the longest English word without repeating letters.
Angry White Men From Wikipedia Too (Score:2)
They want their Wikipedia content back.
How on earth is that supposed to make sense? (Score:2)
I would have thought it is bleeding obvious that whatever is posted on Wiki is freely available to anyone.