9 ms·

EU investigates Google over AI-generated summaries in search results

58 points by hackerbeat 10mo ago

timpera 10mo ago
I'm pretty excited to see how this will develop, especially in the context of "Google Zero". Proving the existence of an anti-competitive effect and quantifying it precisely could be difficult.
- abirch 10mo ago
  I wonder if this will turn into the equivalent of music streaming. Where there's a pot of money that's allocated to different sources. Regardless this is going to negatively impact the current news business model (as do ad blockers and sites that prevent paywalls)
zb3 10mo ago
Compensation to which publishers? To those providing links to SEO spam? If I'm to pay (indirectly) for the content which is used to form the response, we need to match the content that was actually used, not just the content that was sourced, otherwise we'd be rewarding SEO garbage again.
- NewsaHackO 10mo ago
  This is what I don't get. I feel as though the people complaining about this are not the primary source of the information anyway. I wonder if Google already has a way for websites to opt out of the AI mode results (of course, since their sites are not adding any actual new information, there will not be a loss to them). From the search results, it seems like Google has constructed a "Knowledge Graph" LLM which it uses to answer questions in the search results and provide links to sources. How is that different from how every other LLM works? There also seems to be a second issue about Google using YouTuber videos without their consent to train AI, which may be the more pertinent issue the EU is investigating.
akersten 10mo ago
> The European Commission said it would examine whether the firm used data from websites to provide this service - and if it failed to offer "appropriate compensation" to publishers. While the EU wastes their time with things like this, they fall further and further behind the curve, still wondering why no one wants to start a business there.
- jraph 10mo ago
  Behind which curve and how so?
  - DeathArrow 10mo ago
    Behind the curve of big tech businesses generating a lot of GDP and economic growth?
    - Nextgrid 10mo ago
      A lot of said Big Tech is based on industrial-scale fraud and exploitation of the consumer so that a rich few can benefit. Not exactly something to be proud of. (though we too have that in Europe in the form of high taxes, so that a rich few politicians benefit)
    - lores 10mo ago
      Behind the curve of worshipping economic growth over citizen well-being, and behind the curve of being an outright plutocracy, too, so there's that.
- pjmlp 10mo ago
  There are plenty of European businesses, our salaries don't grow on trees.
  - paganel 10mo ago
    Our salaries have certainly not kept up with the pace from across the Atlantic, I'm talking last 10 to 15 years.
Mistletoe 10mo ago
I’m often horrified to follow them down the rabbit hole and see it is a Redditor’s comment. That should terrify you if you have ever used Reddit. Sometimes it is correct, but a lot of times it is very much not right.
- conartist6 10mo ago
  That's accountability-washing in action.
- paganel 10mo ago
  > I’m often horrified to follow them down the rabbit hole and see it is a Redditor’s comment. Genuine question, how are you able to do that? Searching by exact matches with some portions of the AI suggested "response"? Some other method?
  - brainwad 10mo ago
    There are little link icons at the end of each paragraph. They open a list of sources.
cubefox 10mo ago
It seems highly unlikely that AI summaries violate European copyright law. Human summarization is perfectly legal.
- Chabsff 10mo ago
  It's not just a pure matter of law, and looking at it from that perspective is naive. Legacy publishers in general (and a few big ones in particular, like der Spiegel) have been lobbying hard for legislatures to redirect big tech revenue to their failing businesses. The focus on AI here is really just the continuation of that ongoing fight that has been raging for over a decade now. If it wasn't that, it would be some other wedge. I'm not saying Google is squeaky-clean here, far from it. However, it's important to keep in mind that the main drive here is to get publishers paid, not to force Google to be accountable to some specific standards.
- thayne 10mo ago
  I think the argument isn't that it is copyright violation so much as it is anticompetitive. Google is using its monopoly in search to disadvantage its competitors in serving ads (other websites). But on the other hand, when the summaries are accurate (which they aren't always!) they can be beneficial to consumers, so it isn't obviously bad either.
  - cubefox 10mo ago
    > I think the argument isn't that it is copyright violation so much as it is anticompetitive. Google is using its monopoly in search to disadvantage its competitors in serving ads (other websites). But every news website does the same when they summarize the news from other news websites. Which they do all the time.
    - thayne 10mo ago
      But those news sites don't have a monopoly.
      - 10mo ago
OBELISK_ASI 10mo ago
[dead]
sajithdilshan 10mo ago
Why does is feel like EU is creating problems out of nothing just to keep their bureaucrats busy rather than actually doing something worthwhile with tax payers money?
- BDPW 10mo ago
  Appropriate compensation is a non-issue? I have the impression many people jump on the hate-EU train for no other reason than there's many comments reinforcing it. What do you really think about this case in particular? I'm pretty curious where this comes from.
  - sajithdilshan 10mo ago
    Appropriate compensation for what? The summary is generated on the publicly available information.
    - BDPW 10mo ago
      If using data from those websites in a way decreases their visitors or something similar then I think there's an argument to be made for that. I don't know the details to case but just because something is publicly visible doesn't mean that you can just do anything you want with it.
      - sajithdilshan 10mo ago
        There is no guarantee that a website would get a visit if there was no AI summary. Also you can do anything you want with public domain information. That's the whole point of it being public. Otherwise it should be licensed or copyrighted content.
        Timon3 10mo ago
        Almost every news article you come across is copyrighted, and is not public domain.
      - 10mo ago
re-thc 10mo ago
Soon the European Commission will investigate the EU for all its investigations. They might as well just ban all non-EU tech at this point.
boplicity 10mo ago
Google's AI summaries are actively harming quite a lot of people. They're regularly filled with misinformation, but they're presented as facts, complete with references. Many people do not understand the limitations of this technology, and simply believe what they're presented. I'm not convinced that Google understands the limitations, to be honest. The most charitable interpretation I can give of their motivations is that they're terrified of competition from OpenAI, and are trying to present an alternative. Unfortunately, they're presenting a woefully inadequate product. It goes further though, into legitimate questions of copyright, which the tech industry has always fought against. (Take first, deal with it later is the MO.)
ismailmaj 10mo ago
Just in time for the monthly EU bashing
- munksbeer 10mo ago
  Weekly.
  - blitzar 10mo ago
    daily
ZeroConcerns 10mo ago
Although that certainly would have been possible, I've not yet blocked the Gemini AI summaries from the Google search result pages, just because they're so entertaining, in a "sure, bot, you got that entirely wrong" sort of way. And if publisher's rights will be the downfall of that entertainment, I totally get it, but it will be a sad day anyway... (and, quite frankly, my money is still on "libel" for the reason these summaries get nuked in Europe, and it'll be an UK court, not the EU, that triggers this).
hopelite 10mo ago
I still have not gotten anyone to provide a reasonable response to a simple question; if training an AI on some content, how is your reading the same content and then including that in a synthesis of that information along with other information to form your own understanding of the world any different? Alternately, will you start using royalties in perpetuity whenever you talk about some event, because you read an article or a book about that topic once and included something you learned in that article? Basically everything you know, that is even somewhat recent is based on others’ content, do you track and cite every single thing you’ve ever read and send them royalties with every conversation? I’m not trying to defend these big corporations, but for me this is a fundamental question we need to be asking. As consequential as it will be, for me, the answer is that as long as you paid the cost of accessing the content (be it free or a subscription price) while collecting the information that is used to fundamentally transform the information in ways that seem to fall under fair use, then you cannot expect rights, short of full copy/paste plagiarism.
- boplicity 10mo ago
  A training dataset is a document, not a method of processing a document. This type of document regularly gets reproduced and distributed in a commercial environment. Even if the distribution is contained within a large corporation, it still counts as distribution. Should that be allowed within the scope of copyright law? This seems like a legitimate question.
Zufriedenheit 10mo ago
Many news articles have social media posts as sources. Most articles have other articles as sources. And then wikipedia takes the info from the news articles and compiles them. Now google takes all of these and creates summaries again and they have links to original sources in the ai summaries. EU Commission seems very naive and fallen out of time. They are not gonna stuff the AI revolution back into the bottle no matter how hard they try.
- hackerbeat 10mo ago
  Relying on AI to just ‘summarize and link back’ is like expecting a blender to cook a gourmet meal - it’s technically doing something, but the nuance gets lost. Meanwhile, millions of site owners are already watching their traffic drop like ice cubes in a hot sun. The EU isn’t ‘anti-AI,’ they’re just noticing the kitchen is on fire.

58 points by hackerbeat 10mo ago

I'm pretty excited to see how this will develop, especially in the context of "Google Zero". Proving the existence of an anti-competitive effect and quantifying it precisely could be difficult.

abirch 10mo ago
I wonder if this will turn into the equivalent of music streaming. Where there's a pot of money that's allocated to different sources. Regardless this is going to negatively impact the current news business model (as do ad blockers and sites that prevent paywalls)

zb3 10mo ago

Compensation to which publishers? To those providing links to SEO spam? If I'm to pay (indirectly) for the content which is used to form the response, we need to match the content that was actually used, not just the content that was sourced, otherwise we'd be rewarding SEO garbage again.

NewsaHackO 10mo ago
This is what I don't get. I feel as though the people complaining about this are not the primary source of the information anyway. I wonder if Google already has a way for websites to opt out of the AI mode results (of course, since their sites are not adding any actual new information, there will not be a loss to them). From the search results, it seems like Google has constructed a "Knowledge Graph" LLM which it uses to answer questions in the search results and provide links to sources. How is that different from how every other LLM works? There also seems to be a second issue about Google using YouTuber videos without their consent to train AI, which may be the more pertinent issue the EU is investigating.

akersten 10mo ago

> The European Commission said it would examine whether the firm used data from websites to provide this service - and if it failed to offer "appropriate compensation" to publishers. While the EU wastes their time with things like this, they fall further and further behind the curve, still wondering why no one wants to start a business there.

jraph 10mo ago
Behind which curve and how so?
- DeathArrow 10mo ago
  Behind the curve of big tech businesses generating a lot of GDP and economic growth?
  - Nextgrid 10mo ago
    A lot of said Big Tech is based on industrial-scale fraud and exploitation of the consumer so that a rich few can benefit. Not exactly something to be proud of. (though we too have that in Europe in the form of high taxes, so that a rich few politicians benefit)
  - lores 10mo ago
    Behind the curve of worshipping economic growth over citizen well-being, and behind the curve of being an outright plutocracy, too, so there's that.
pjmlp 10mo ago
There are plenty of European businesses, our salaries don't grow on trees.
- paganel 10mo ago
  Our salaries have certainly not kept up with the pace from across the Atlantic, I'm talking last 10 to 15 years.

Mistletoe 10mo ago

I’m often horrified to follow them down the rabbit hole and see it is a Redditor’s comment. That should terrify you if you have ever used Reddit. Sometimes it is correct, but a lot of times it is very much not right.

conartist6 10mo ago
That's accountability-washing in action.
paganel 10mo ago
> I’m often horrified to follow them down the rabbit hole and see it is a Redditor’s comment. Genuine question, how are you able to do that? Searching by exact matches with some portions of the AI suggested "response"? Some other method?
- brainwad 10mo ago
  There are little link icons at the end of each paragraph. They open a list of sources.

cubefox 10mo ago

It seems highly unlikely that AI summaries violate European copyright law. Human summarization is perfectly legal.

Chabsff 10mo ago
It's not just a pure matter of law, and looking at it from that perspective is naive. Legacy publishers in general (and a few big ones in particular, like der Spiegel) have been lobbying hard for legislatures to redirect big tech revenue to their failing businesses. The focus on AI here is really just the continuation of that ongoing fight that has been raging for over a decade now. If it wasn't that, it would be some other wedge. I'm not saying Google is squeaky-clean here, far from it. However, it's important to keep in mind that the main drive here is to get publishers paid, not to force Google to be accountable to some specific standards.
thayne 10mo ago
I think the argument isn't that it is copyright violation so much as it is anticompetitive. Google is using its monopoly in search to disadvantage its competitors in serving ads (other websites). But on the other hand, when the summaries are accurate (which they aren't always!) they can be beneficial to consumers, so it isn't obviously bad either.
- cubefox 10mo ago
  > I think the argument isn't that it is copyright violation so much as it is anticompetitive. Google is using its monopoly in search to disadvantage its competitors in serving ads (other websites). But every news website does the same when they summarize the news from other news websites. Which they do all the time.
  - thayne 10mo ago
    But those news sites don't have a monopoly.
    - 10mo ago

OBELISK_ASI 10mo ago

[dead]

sajithdilshan 10mo ago

Why does is feel like EU is creating problems out of nothing just to keep their bureaucrats busy rather than actually doing something worthwhile with tax payers money?

BDPW 10mo ago
Appropriate compensation is a non-issue? I have the impression many people jump on the hate-EU train for no other reason than there's many comments reinforcing it. What do you really think about this case in particular? I'm pretty curious where this comes from.
- sajithdilshan 10mo ago
  Appropriate compensation for what? The summary is generated on the publicly available information.
  - BDPW 10mo ago
    If using data from those websites in a way decreases their visitors or something similar then I think there's an argument to be made for that. I don't know the details to case but just because something is publicly visible doesn't mean that you can just do anything you want with it.
    - sajithdilshan 10mo ago
      There is no guarantee that a website would get a visit if there was no AI summary. Also you can do anything you want with public domain information. That's the whole point of it being public. Otherwise it should be licensed or copyrighted content.
      - Timon3 10mo ago
        Almost every news article you come across is copyrighted, and is not public domain.
    - 10mo ago

re-thc 10mo ago

Soon the European Commission will investigate the EU for all its investigations. They might as well just ban all non-EU tech at this point.

boplicity 10mo ago

Google's AI summaries are actively harming quite a lot of people. They're regularly filled with misinformation, but they're presented as facts, complete with references. Many people do not understand the limitations of this technology, and simply believe what they're presented. I'm not convinced that Google understands the limitations, to be honest. The most charitable interpretation I can give of their motivations is that they're terrified of competition from OpenAI, and are trying to present an alternative. Unfortunately, they're presenting a woefully inadequate product. It goes further though, into legitimate questions of copyright, which the tech industry has always fought against. (Take first, deal with it later is the MO.)

ismailmaj 10mo ago

Just in time for the monthly EU bashing

munksbeer 10mo ago
Weekly.
- blitzar 10mo ago
  daily

ZeroConcerns 10mo ago

Although that certainly would have been possible, I've not yet blocked the Gemini AI summaries from the Google search result pages, just because they're so entertaining, in a "sure, bot, you got that entirely wrong" sort of way. And if publisher's rights will be the downfall of that entertainment, I totally get it, but it will be a sad day anyway... (and, quite frankly, my money is still on "libel" for the reason these summaries get nuked in Europe, and it'll be an UK court, not the EU, that triggers this).

hopelite 10mo ago

I still have not gotten anyone to provide a reasonable response to a simple question; if training an AI on some content, how is your reading the same content and then including that in a synthesis of that information along with other information to form your own understanding of the world any different? Alternately, will you start using royalties in perpetuity whenever you talk about some event, because you read an article or a book about that topic once and included something you learned in that article? Basically everything you know, that is even somewhat recent is based on others’ content, do you track and cite every single thing you’ve ever read and send them royalties with every conversation? I’m not trying to defend these big corporations, but for me this is a fundamental question we need to be asking. As consequential as it will be, for me, the answer is that as long as you paid the cost of accessing the content (be it free or a subscription price) while collecting the information that is used to fundamentally transform the information in ways that seem to fall under fair use, then you cannot expect rights, short of full copy/paste plagiarism.

boplicity 10mo ago
A training dataset is a document, not a method of processing a document. This type of document regularly gets reproduced and distributed in a commercial environment. Even if the distribution is contained within a large corporation, it still counts as distribution. Should that be allowed within the scope of copyright law? This seems like a legitimate question.

Zufriedenheit 10mo ago

Many news articles have social media posts as sources. Most articles have other articles as sources. And then wikipedia takes the info from the news articles and compiles them. Now google takes all of these and creates summaries again and they have links to original sources in the ai summaries. EU Commission seems very naive and fallen out of time. They are not gonna stuff the AI revolution back into the bottle no matter how hard they try.

hackerbeat 10mo ago
Relying on AI to just ‘summarize and link back’ is like expecting a blender to cook a gourmet meal - it’s technically doing something, but the nuance gets lost. Meanwhile, millions of site owners are already watching their traffic drop like ice cubes in a hot sun. The EU isn’t ‘anti-AI,’ they’re just noticing the kitchen is on fire.