I feel like this is a headline that does a good job of capturing the heart of the issue.
I’ve tried to evaluate how useful ai is so I can at least get a sense for how much harm it will do, as its WAY more potentially harmful if its useful enough to actually seen any organic adoption. And my experience has been that it can be incredibly helpful, particular for finding things online that are out there somewhere but that search engines are just useful for finding or consolidating into one place (it feels like a major element of that is just how ass search engines are. Part of me feels like google is comfortable allowing their traditional search experience to languish because it makes ai more necessary)
But the heart of the problem is that those upsides for users feel largely incidental to the way companies are forcing adoption down everyone’s throats, and the massive harms it caries are all feel like fairly structural elements of why anyone cares about the technology as a product to sell, or elements of how the technology works.
The intellectual property theft and labor exploitation, environmental harms, skyrocketing utility and noise pollution for people near data centers, harm to societies ability to think critically, stochastic parrot tendency to spread disinformation or misunderstanding, and supercharged surveillance capitalism are all things that are fairly intrinsic. Some of them can be avoided with a local model (which I would guess may make the model way worse functionally, but at some point I need to try options in that space so I understand where things are at)
At the end of the day, the reason we’re burning ludicrous amounts of money (and electricity) to prop up this technologies adoption before its even remotely monetizable is NECCISARILY because it has the potential to rob people of their employment by stealing labor other people have done to create a commercial product. Thats the only aspect of it that makes it a worthwhile investment for companies. So companies will move heaven and earth to be the ones who hold the most competitive version, and force people to become dependent on this tech so that it is normalized to the point we can’t criticize it and extricate it from our society once its been successfully woven in.
So its fun to laugh when its really dumb and its lack of actual understanding under-the-hood is on full display, the reality is that sentiment drastically underrepresents the amount of harm that it might do to society. Its not that its never helpful, its that the reason its available to us is for the explicit purpose of enabling corporations to do harm to our world for profit
And given the way it impacts user’s thinking, its also harmful on the individual scale. Frankly I can tell its bad for my head, and so I do my best to limit use. But the fact that sometimes its the most effective way to find certain things makes it so easy to come back to, and so easy to just keep asking questions and getting easy polished answers. It feels slightly addicting in nature.
Thats the heart of it. Its more harmful than helpful.
Part of me feels like google is comfortable allowing their traditional search experience to languish because it makes ai more necessary
Oh, it’s so much worse than that, Google intentionally made it worse around 2019 so that people would do multiple searches and scroll to second pages, thus increasing the amount of ad impressions and user time spent on the site. There were several email back and forth between the head of search and head of advertising, with the head of search adamantly refusing to implement the changes. Eventually he ended up leaving despite having been at the company from the beginning, due to this disagreement. The head of advertising during this? He’s the head of search now.
Replacing search altogether with AI summarization is just a continuation of that, instead of delaying customers going to other sites, prevent them from going to them all together.
That was sundar pichai? Or someone else? Do you know what terms I’d need to look up to find sources for that? Thats information I’d absolutely like to include when talking about this with people, thank you for filling me in, I had no idea about that and it adds a lot of substance to that point
Edit: from the provided link the person who killed google search was Prabhakar Raghavan, who was apparently responsible for yahoo search over the period in which it fell precipitously off a cliff. He was put in that role by sundar pichai, replacing someone who explicitly cared about search and was deeply important to search development and had been there since the beginning
Do you want to know what Prabhakar Raghavan’s old job was? What Prabhakar Raghavan, the new head of Google Search, the guy that has run Google Search into the ground, the guy who is currently destroying search, did before his job at Google?
He was the head of search for Yahoo from 2005 through 2012 — a tumultuous period that cemented its terminal decline, and effectively saw the company bow out of the search market altogether. His responsibilities? Research and development for Yahoo’s search and ads products
So not Sundar pichai, but the article does explain that he was specifically picked to become the new head of search by sundar pichai, and that Sundar used to work for McKinsey Consulting, who were responsible for the business strategy that fueled the opioid crisis, among lots of other things. Super dense, but well worth the read!
Thank you so much!! There were some details I wanted to clarify for people after reading so I edited my previous reply so folks can see a summary of some of the relevant points
I appreciate you getting me a link! This will be very helpful in explaining some of the rot in the search experience and adds a lot of additional weight to my suspicion that making search worse may also be happening to push ai adoption
i looked up the story I had read a while back, and sundar was not the head of ads, but had just become the ceo after having been on the board of directors, and the issue was kicked up to him when Ben Gomes refused implement changes that Prabhakar Raghavan of ads wanted. later Gomes left the company and was replaced by Raghavan as head of search.
This all coming from internal emails and memos that were released as part of the anti trust case summarized in this piece which does a very good job of sourcing individual claims by pointing to specific emails and memos, although, the author is very… passionate in their coverage.
Very rarely I see the argument that apart from being harmful directly, it’s also a GIGANTIC surveilance tool, and the most vulnerable people trust it with their inner thougths, which is to me scarier that ‘just’ copyrights.
Absolutely agree, the data collection feels like even a step beyond the already horrifying amount of people’s personal information we’ve normalized scraping directly into the pockets of corporations
Yeah I need to try those models at some point so I understand them a bit better. Not sure how my laptop will like them though, or if I’ll be functionality limited by compute to the point that they’re less than useful.
To clarify: Jan, Ollama, Alpaca, etc. are frameworks, onto which you can download and chat with different LLMs. They themselves are not LLMs. But yeah, you should probably have, like, a 2080+ or something…
I see it as intrinsic to the reason the technology is being invested in and pushed into everything. Ai is wildly unmonetizable outside of data collection and replacing jobs
A similar but slightly different thing happened with crypto which once upon a time was about the idea of currency that didnt pass through a centralized controlling entities (the goverment, payment processors, etc) and thats still relevant today as we descend into authoritarianism and corporations censor legal businesses because they engage with sexuality. And yet crypto as a space and a technology is mired in abuse and exploitation- why?
Because its deeply profitable for harmful things (crime, scams, wealth transfer via a hollow investment vehicle). The good it can do is not the driving force behind it as an entity. That potentially even more true about ai given it is a commercial product built by massive companies at an enormous loss. And ai has the potential to have a much more profound impact on the world than crypto, rewriting deeply important social contracts in the image of corporate profit.
The human cultural norms around creating and sharing intellectual and artistic works that have been accepted foundations of human society are kind of being dissolved so that companies can take value from those works that they didn’t compensate anyone for.
I see where you are coming from, even if I don’t see a direct link between AI and hyper surveillance in your text. Personally I wager people will not only upload their holiday photos to the big corporate but share their most inner feelings with it, volunteerly…
And, like facebook did, they’ll create/generate “shadow persons” descriptors for people who are not using AI, by just using people around them.
On a side note, art cannot be made by machines, it’s intrinsic to the human mind, an expression of it. AI can rehash and redistribute it though, which is only bad for artists doing business (so that’s the downside, not excusing it but machines have done that like forever, again not excusing anything).
If we finally distributed some of the wealth automation generates we could all create art (and machines could even distribute it), but that’s dreamland for now of course.
Yeah, in a perfect world artists would not have to pay bills to survive, though I don’t know that thats really an art specific problem. If a company finds a novel way to exploit steel workers, that does happen within the backdrop of those people deserving to survive even if they can’t produce monetary value for someone else, but it feels a bit like it misses the immediate actionable harm, in favor of the less easily addressed bigger picture problem
But you’re not wrong, the underlying structural issue is people’s ability to survive being predicated on ones commercial viability, and the most commercially viable path is always rent seeking behavior and hiring people you will pay less than the ammount of value they create
With respect to surveillance, even in my limited usage I’ve shared a hugely valuable ammount of information from the perspective of understanding me as a consumer. Though, I do very much see a huge implication for regular old surveillance as well.
All companies in the us right now are operating within an environment where they need to kiss the ring, and where kissing the ring passionately enough can be competitively advantageous to them as a company, and beneficial to their bottom line. If the government comes knocking and asks an ai company to give them data, whether to create a case against a specific individual, or for access to ongoing usage activity/data sharing with federal agencies, they have very good reason to consider that course. And my understanding is that they are free to just say yes as long as their TOS doesnt preclude it.
Replace AI with something like Flex Tape, and you really get a sense of how stupid the push to get AI into everything is. Flex Tape is great, but I don’t need it when ordering a pizza ffs.
I feel like this is a headline that does a good job of capturing the heart of the issue.
I’ve tried to evaluate how useful ai is so I can at least get a sense for how much harm it will do, as its WAY more potentially harmful if its useful enough to actually seen any organic adoption. And my experience has been that it can be incredibly helpful, particular for finding things online that are out there somewhere but that search engines are just useful for finding or consolidating into one place (it feels like a major element of that is just how ass search engines are. Part of me feels like google is comfortable allowing their traditional search experience to languish because it makes ai more necessary)
But the heart of the problem is that those upsides for users feel largely incidental to the way companies are forcing adoption down everyone’s throats, and the massive harms it caries are all feel like fairly structural elements of why anyone cares about the technology as a product to sell, or elements of how the technology works.
The intellectual property theft and labor exploitation, environmental harms, skyrocketing utility and noise pollution for people near data centers, harm to societies ability to think critically, stochastic parrot tendency to spread disinformation or misunderstanding, and supercharged surveillance capitalism are all things that are fairly intrinsic. Some of them can be avoided with a local model (which I would guess may make the model way worse functionally, but at some point I need to try options in that space so I understand where things are at)
At the end of the day, the reason we’re burning ludicrous amounts of money (and electricity) to prop up this technologies adoption before its even remotely monetizable is NECCISARILY because it has the potential to rob people of their employment by stealing labor other people have done to create a commercial product. Thats the only aspect of it that makes it a worthwhile investment for companies. So companies will move heaven and earth to be the ones who hold the most competitive version, and force people to become dependent on this tech so that it is normalized to the point we can’t criticize it and extricate it from our society once its been successfully woven in.
So its fun to laugh when its really dumb and its lack of actual understanding under-the-hood is on full display, the reality is that sentiment drastically underrepresents the amount of harm that it might do to society. Its not that its never helpful, its that the reason its available to us is for the explicit purpose of enabling corporations to do harm to our world for profit
And given the way it impacts user’s thinking, its also harmful on the individual scale. Frankly I can tell its bad for my head, and so I do my best to limit use. But the fact that sometimes its the most effective way to find certain things makes it so easy to come back to, and so easy to just keep asking questions and getting easy polished answers. It feels slightly addicting in nature.
Thats the heart of it. Its more harmful than helpful.
Oh, it’s so much worse than that, Google intentionally made it worse around 2019 so that people would do multiple searches and scroll to second pages, thus increasing the amount of ad impressions and user time spent on the site. There were several email back and forth between the head of search and head of advertising, with the head of search adamantly refusing to implement the changes. Eventually he ended up leaving despite having been at the company from the beginning, due to this disagreement. The head of advertising during this? He’s the head of search now.
Replacing search altogether with AI summarization is just a continuation of that, instead of delaying customers going to other sites, prevent them from going to them all together.
That was sundar pichai? Or someone else? Do you know what terms I’d need to look up to find sources for that? Thats information I’d absolutely like to include when talking about this with people, thank you for filling me in, I had no idea about that and it adds a lot of substance to that point
Edit: from the provided link the person who killed google search was Prabhakar Raghavan, who was apparently responsible for yahoo search over the period in which it fell precipitously off a cliff. He was put in that role by sundar pichai, replacing someone who explicitly cared about search and was deeply important to search development and had been there since the beginning
So not Sundar pichai, but the article does explain that he was specifically picked to become the new head of search by sundar pichai, and that Sundar used to work for McKinsey Consulting, who were responsible for the business strategy that fueled the opioid crisis, among lots of other things. Super dense, but well worth the read!
Here you go
Thank you so much!! There were some details I wanted to clarify for people after reading so I edited my previous reply so folks can see a summary of some of the relevant points
I appreciate you getting me a link! This will be very helpful in explaining some of the rot in the search experience and adds a lot of additional weight to my suspicion that making search worse may also be happening to push ai adoption
i looked up the story I had read a while back, and sundar was not the head of ads, but had just become the ceo after having been on the board of directors, and the issue was kicked up to him when Ben Gomes refused implement changes that Prabhakar Raghavan of ads wanted. later Gomes left the company and was replaced by Raghavan as head of search.
This all coming from internal emails and memos that were released as part of the anti trust case summarized in this piece which does a very good job of sourcing individual claims by pointing to specific emails and memos, although, the author is very… passionate in their coverage.
Very rarely I see the argument that apart from being harmful directly, it’s also a GIGANTIC surveilance tool, and the most vulnerable people trust it with their inner thougths, which is to me scarier that ‘just’ copyrights.
Absolutely agree, the data collection feels like even a step beyond the already horrifying amount of people’s personal information we’ve normalized scraping directly into the pockets of corporations
This is why FOSS is the safest way to go, with Ollama, Alpaca, etc. You’d need a beefy rig to run such models, though.
Yeah I need to try those models at some point so I understand them a bit better. Not sure how my laptop will like them though, or if I’ll be functionality limited by compute to the point that they’re less than useful.
To clarify: Jan, Ollama, Alpaca, etc. are frameworks, onto which you can download and chat with different LLMs. They themselves are not LLMs. But yeah, you should probably have, like, a 2080+ or something…
Righ, ive been corrected about that before and keep forgetting, thank you lol.
Yeah I have integrated graphics, I don’t think my laptop is gonna like that attempt at testing…
Why do you think supercharged surveillance capitalism is intrinsic?
I see it as intrinsic to the reason the technology is being invested in and pushed into everything. Ai is wildly unmonetizable outside of data collection and replacing jobs
A similar but slightly different thing happened with crypto which once upon a time was about the idea of currency that didnt pass through a centralized controlling entities (the goverment, payment processors, etc) and thats still relevant today as we descend into authoritarianism and corporations censor legal businesses because they engage with sexuality. And yet crypto as a space and a technology is mired in abuse and exploitation- why?
Because its deeply profitable for harmful things (crime, scams, wealth transfer via a hollow investment vehicle). The good it can do is not the driving force behind it as an entity. That potentially even more true about ai given it is a commercial product built by massive companies at an enormous loss. And ai has the potential to have a much more profound impact on the world than crypto, rewriting deeply important social contracts in the image of corporate profit.
The human cultural norms around creating and sharing intellectual and artistic works that have been accepted foundations of human society are kind of being dissolved so that companies can take value from those works that they didn’t compensate anyone for.
I see where you are coming from, even if I don’t see a direct link between AI and hyper surveillance in your text. Personally I wager people will not only upload their holiday photos to the big corporate but share their most inner feelings with it, volunteerly…
And, like facebook did, they’ll create/generate “shadow persons” descriptors for people who are not using AI, by just using people around them.
On a side note, art cannot be made by machines, it’s intrinsic to the human mind, an expression of it. AI can rehash and redistribute it though, which is only bad for artists doing business (so that’s the downside, not excusing it but machines have done that like forever, again not excusing anything).
If we finally distributed some of the wealth automation generates we could all create art (and machines could even distribute it), but that’s dreamland for now of course.
Yeah, in a perfect world artists would not have to pay bills to survive, though I don’t know that thats really an art specific problem. If a company finds a novel way to exploit steel workers, that does happen within the backdrop of those people deserving to survive even if they can’t produce monetary value for someone else, but it feels a bit like it misses the immediate actionable harm, in favor of the less easily addressed bigger picture problem
But you’re not wrong, the underlying structural issue is people’s ability to survive being predicated on ones commercial viability, and the most commercially viable path is always rent seeking behavior and hiring people you will pay less than the ammount of value they create
With respect to surveillance, even in my limited usage I’ve shared a hugely valuable ammount of information from the perspective of understanding me as a consumer. Though, I do very much see a huge implication for regular old surveillance as well.
All companies in the us right now are operating within an environment where they need to kiss the ring, and where kissing the ring passionately enough can be competitively advantageous to them as a company, and beneficial to their bottom line. If the government comes knocking and asks an ai company to give them data, whether to create a case against a specific individual, or for access to ongoing usage activity/data sharing with federal agencies, they have very good reason to consider that course. And my understanding is that they are free to just say yes as long as their TOS doesnt preclude it.
(Sorry, I’m really long winded lol 😅)
Replace AI with something like Flex Tape, and you really get a sense of how stupid the push to get AI into everything is. Flex Tape is great, but I don’t need it when ordering a pizza ffs.