This framing confuses me. Google has shifted its product to be a search engine with AI first. The request here is effectively that Google built and maintain a second version of their product with no front facing llm, and Google has declined to, correct?
It isn't like there is some second version of Google that Google is withholding from schools. However, there are search engines which don't use AI. Switching to those would probably do more than protests to get Google to switch its product roadmap.
Couldn't they just turn off the generative summary and go back to showing search results instead? It doesn't seem like asking for too much additional from one of the world's largest and most capable companies.
Technically, perhaps, but the future is in LLM answers. The old style search is likely to rot as Google invests everything in the AI mode.
Or another perspective: Nobody expects to be able to ask Google for a version without ads, even though ads are annoying, because they know Google makes most of its revenue from search ads. Why do they then presume to dictate that Google give up on their strategic priority of pivoting to AI? It's equally absurd. Thankfully we have long had a solution for this: client-side blocking.
SEO didn't force Google to ignore half your search terms or give you results for thematically or semantically similar search terms that are ultimately irrelevant to what you actually searched for.
That was all the doing of employees at Google. SEO did not have even the tiniest iota of relevance for Google's discretionary choice to make search results worse.
But even though they are worse, they are not yet useless.
Only if chrome supports the required extensions, which with a drop of manifest v2 they only do half assed. School boards can also filter at the network level but that's very hard to get right and can cause a whole new set of problems
Ads pay for the product, and AI doesn't. It's absurd to ask them to burn money, it's not absurd to want a toggle for a new feature that causes problems.
And they'll fight back against blocking.
(Also asking for a version without ads is reasonable if you're willing to pay a modest fee for it. So that's not absurd either.)
Sure they do. Google is under no obligation to bend over backwards to make users of its (free) product all happy. It's a competitive market, if you don't like what Google offers, don't use it. What Google offered schools is strictly more control than before, and they complained. Lesson: offer less control?
No, Google still has a search engine that returns pages of results like it is 1999. The Gemini iframe (or whatever) sits on top of those pages of results.
This is just willfully misunderstanding what the schools want. It's very obvious they want the summary/chatbot interface to go away, and to only show links that are shown below the chatbox instead. Nobody is asking anyone to change the ranking.
This is not a particularly big ask. If it truly is lost science to not add a chatbox above the links, they could hide it it via the CSS layer if needs be. If nobody remains at google that knows how to hide a div with CSS, half the people in this comment thread could help them with that.
Their requirements are broader than the user interface. Some boards have policies to prevent students from using AI, or using student content for AI training. In other words, there can be no read or write pathways that include AI.
There is no reasonable way for Google to do that (that they could affordably sell to the schools).
>> This is just willfully misunderstanding what the schools want. It's very obvious they want the summary/chatbot interface to go away, and to only show links that are shown below the chatbox instead. Nobody is asking anyone to change the ranking.
> I think you are projecting your own desires onto the schools, and if that were actually what the schools wanted, it would not be an issue
It's pretty obvious you are wrong and the GP is correct.
If Google's ranking was done by AI for the last "12-15 years", the schools didn't care. But now Google adds a fucking chatbot that the kids can ask to do their homework, and that's when the schools start caring. It's pretty obvious the chatbot and AI overviews are what they care about.
Certainly it's a fact that the schools didn't care in the past, but I think that's mostly a result of AI not being in the spotlight then, and there wasn't a broad anti-AI faction then.
From TFA:
> Attendees of all ages held signs reading, “NO AI IN MY SCHOOL!”, “A.I. Gets an F,” “Stop stealing our imaginations,” and “SCHOOL Librarians = THE BEST Search Engines.”
I'm pretty sure these people would not be okay with plain search if they understood the extent to which it uses and has used AI.
(Also, I suspect you and other commenters have an implicit belief that the AI plain search uses and what the AI box does are fundamentally different, and I doubt that. The search algo is going to incorporate new tech as it comes along, and it's unlikely there is zero LLM involvement.)
> Certainly it's a fact that the schools didn't care in the past, but I think that's mostly a result of AI not being in the spotlight then, and there wasn't a broad anti-AI faction then.
> From TFA:
>> Attendees of all ages held signs reading, “NO AI IN MY SCHOOL!”, “A.I. Gets an F,” “Stop stealing our imaginations,” and “SCHOOL Librarians = THE BEST Search Engines.”
> I'm pretty sure these people would not be okay with plain search if they understood the extent to which it uses and has used AI.
I doubt it, and if they did, it'd probably be due to misunderstanding where they conflate LLMs with pre-LLM AI technology (e.g. the "12-15 years" tech that was used for Google Search ranking). The emotions are pretty clearly being driven by the current thing (LLMs and similar models).
I think you're making the mistake of interpreting their statements too literally, using definitions and interpretations that they simply don't share, to falsely twist their statements into embarrassing contradictions. You can use the word "AI" and they can use the word "AI" and you can both mean different things (or have some conceptual overlap but with significant differences).
> (Also, I suspect you and other commenters have an implicit belief that the AI plain search uses and what the AI box does are fundamentally different, and I doubt that. The search algo is going to incorporate new tech as it comes along, and it's unlikely there is zero LLM involvement.)
When they say AI, they are almost certainly referring to the chatbots, the AI overviews, etc. and excluding search, which has no user-obvious AI component. Being "but ackchyually" to try to turn their position on its head can be pretty disrespectful.
School board's set policy at a high level. If they pass a resolution "no student facing AI" -- now Google has to turn a lot of things off that were on before.
What do people gain from being obtuse? You want to pretend to be confused even though you're probably smart enough not to be? It's very simple, people want the traditional list of links search results and not LLM-written overviews. It's not like that has been thrown away or isn't still in production.
The move is 100% punitive on the part of Google. From the recent CNN article:
"Google is rolling out new controls for schools to limit students’ access to the company’s AI tools. But if schools opt out of AI altogether, they’ll also lose access to Google Search."
They are saying that if schools would like to disable all of the AI tooling in the Chromebooks used by students then Google will also disable access to search.
>They are saying that if schools would like to disable all of the AI tooling in the Chromebooks used by students then Google will also disable access to search.
THAT is a clearer answer than any I received. If the demand is "let us turn off AI features generally, but keep search with AI" then that seems like an extremely reasonable request. I'm not sure if that is the request but it's not outlandish vs. maintaining a contractually provisioned no AI search engine version.
What they are aiming towards is irrelevant. Your argument could just as well be interpreted as "Google is trying to mandate students rely on AI summaries so Google can one day stop linking to websites that are not owned by Google".
If anything, this is an argument in favor of the schools and against Google.
If Google wasn't still linking to external websites then you might have a point. But they are. So what Google wants to "nudge" students to do is irrelevant when schools - the entity actually empowered to teach students - are asking them to hide an iframe when students are using search.
Until the day Google stops linking to the rest of the Internet, that will be the most trivial of asks.
> The request here is effectively that Google built and maintain a second version of their product with no front facing llm, and Google has declined to, correct?
Google already has a web search with no front-facing LLM, and you can get it by selecting "Web" from a search results page, or by including udm=web in a constructed search URL like so:
Somehow, it's impossible for Google to provide schools with the no-AI-summary search engine they already provide! My personal opinion of Google's behavior is in the example URL that I gave.
Or by including -whatever at the end of your search query... It's not a specific "-ai" parameter, AI isn't yet trained to answer negative queries so if Google sees a negative query it disables AI overview.
Are the schools explicitly asking for "no front-facing LLM"/"no AI-summary", or are they asking for "no AI"? Those are two very different things. Separate from the LLM-generated summaries, the basic Google Search results ranking have used various AI techniques for well over a decade. It should be easy for Google to turn off the summaries for specific organizations, but it's infeasible for them to go back to the pure Pagerank results from the 2000s.
> Starting next year, Google said, its A.I.-related products, including Search and its Gemini app, will be turned off automatically for schools, and administrators can choose to turn those A.I. tools on. But there will be no version of search available without A.I. chatbot features, meaning schools that want to forgo A.I. will lose access to an important learning tool.
It's specifically chatbots that Google is saying they won't allow schools to disable without disabling search as a whole.
> Are the schools explicitly asking for "no front-facing LLM"/"no AI-summary", or are they asking for "no AI"? Those are two very different things. Separate from the LLM-generated summaries, the basic Google Search results ranking have used various AI techniques for well over a decade
The first one. They've made that clear, and also, evidence over the past decade+ indicates they aren't concerned with the second one. And why would they be?
abiding by a "no AI" policy means not presenting or contributing to AI content. AI is not limited to AI chatbots (like chatgpt). it would mean any read or write path that leads to AI training or inference.
Google search content reads and writes AI pathways, even on "normal" search (aka Web search tab)
Do you have evidence of that? Because personally i would be very surprised that google is maintaining a non-AI version just for kicks and not using it anywhere. That is just not how tech companies operate.
They have to maintain a web search engine regardless, because the LLM needs it. It's not just hallucinating facts from its weights. (I mean, occasionally it will do that, but usually it's based on search results.)
This isn't about maintaining the back end. It's about how the search results are presented.
> It's not just hallucinating facts from its weights. (I mean, occasionally it will do that
My experience is the Google AI block at the top of each search has very frequent hallucinations presumably caused by the fast model required so users don't scroll past it before it's finished loading.
> Google has shifted its product to be a search engine with AI first. The request here is effectively that Google built and maintain a second version of their product with no front facing llm
Guys, we're all programmers here. We all know this is the kind of bullshit engineers say when they don't want to do something. We don't have to fool each other.
I don't know how Google Search works, but I can guarantee you that they don't have LLM responses intrinsically linked to search results. If their LLM services go down, I'm pretty sure they can still serve search results. All that schools want is to have that flag exposed in the account/organization settings page.
This is the same type of bullshit Microslop tried to pull with Windows XP: saying that they can't unhook Internet Explorer from Windows. The EU forced them and it turned out it was possible all of a sudden.
Saying "maintain a second version" is so hyperbolic. All that's being asked for is making a flag public. A single Google engineer should be able to code it (even without AI) and deploy it in 2 hours worldwide.
Every widget on Google’s search page is an independent component maintained by different teams. There are thousands. Each one can be configured to trigger or not based on thousands of different switches (environment, product config, experiments, etc.)
It’s dead simple to turn off singular components in specific configs.
Right now, but once you contractually sell a "no AI" product then you need to maintain a version which has "no AI". The existing of a link form which has no AI presently is very different from a contractual commitment to institutional buyers who will complain loudly and sue if the contractual conditions are not met.
No you just let someone pay to be on an IP address list and when handling a request, if the client IP is on that list, you skip over the section of the code that generates the AI summary.
And if they get around the restriction or find a way to ai mode is Google contractually liable?
To feel confidant in the delivery I would expect Google to need to build and maintain a no-ai search engine, indefinitely, even if their current search experience becomes more ai-centric.
This was my point. I think Google knows their own project roadmap and doesn't want to contractually build and maintain a legacy no-ai version of search.
Google sold this product to many schools during the pandemic. In my opinion, it is an epic fail on their part that they have ignored the needs of this market segment. That they now expect them to come up with their own solution to a problem that Google has created seems unreasonable to me.
This is a trivial change for Google. I worry that the rationale is that if they force-feed the AI summaries to children then those children are more likely to grow up into customers willing to purchase LLM products or be generally more willing to accept LLM content. Which is kind of gross.
This is the dumbest strawman argument, and I can't believe people are still stooping to it to carry water for Google.
Schools do not care whether Google uses AI under the hood. They do not care whether Google sorts its results using AI.
They do want Gemini to stop answering student questions when they're typed into Google, and to stop giving summaries at the top.
The dumbest, working solution would be a one-line CSS fix to hide the Gemini box at the top of search results for schools. A smarter solution is easily within the grasp of Google.
Exactly. If the teacher asks the student a question and the computer does it for them, then what's the point? Your homework is to type this into Google and copy and paste what comes out.
If customers are literally asking for a certain customization like this, I don't see why it's a crazy ask. I don't hate the AI features in Google but I think it makes sense to have a setting somewhere to disable it.
The request is to make the visible bits (AI mode, AI overview) hidden. Schools don’t care if Google is using classical PageRank or AI-based ranking under the hood.
Here is the part you are missing: Google sold Chromebooks and G Suite to schools for the last 15 years under the assurance that these things will improve education. Schools are now telling Google that Gemini is a serious problem for cheating and coursework, and Google's response is that they can't be bothered to put in a feature flag for this new feature that is causing problems. For institutions that have collectively spent tens of billions on locked down Google hardware, that is a thumb to the eye.
The most generous interpretation of Google's response is that Google's product suite is not aligned with the goals of educational institutions and that Google isn't willing to put in the work to get it there. The actual story of course is that Google is a monopoly that lobbied hard to get schools locked into their ecosystem and now they are pushing their weight around to try to get kids hooked on their AI early.
> Most spend 12 years in school and come out with no marketable skills. Something could be wrong with that.
You’re not wrong, but schools are only partly to blame for that. Plenty of other factors out there. Don’t make the mistake of blaming schooling for a broken economic system and what are fundamentally social class issues.
They are wrong. 17 year olds don't need marketable skills. They need experience, and an opportunity to build on their their substrate of general knowledge and deeply study one or two areas that interest them.
If school isn't delivering that, then yes, they are broken. But no-where in schooling is the requirement to deliver someone with "marketable skills".
Aside from mathematics, which I use professionally on a daily basis, the most impactful lessons I learned in high school were in home economics. I’m not sure that’s not even taught anymore, which is a shame. I learned how to run a kitchen and keep it clean, how to plan a personal budget, how to do basic clothing repair, sewing, and so on. It all sounds mundane but I grew up in a house where some of those things (cleaning in particular) did not happen and it was a real revelation to me. I don’t think 14-year-old-me would have chosen to take those classes, but I am so glad that I did.
> Most spend 12 years in school and come out with no marketable skills. Something could be wrong with that.
There's nothing wrong with that. That's the base for learning "marketable skills". 12 years gives you a solid grounding in basics, and you start to get an idea of what you're maybe good at, or interested in, and then you get the opportunity to diversify, to take that basic learning to university or college and study one or two specific things, or get some kind of entry-level/junior position.
17 year olds don't need "marketable skills". They need encouragement, opportunity, and experience.
And anyway, fuck your "marketable skills". Knitting is not a marketable skill, but you know what, it shows skill, application, and dedication. Not everything learned in school is marketable, and there's nothing wrong with that and everything wrong with thinking otherwise.
Foundational education shouldn't be up for debate. It should be curriculums designed by experts to give us a foundation of knowledge to make responsible choices about what comes next. It's why you don't get to pick classes till high school, and certain types of courses are still required in college. Sometimes you need to know stuff you aren't motivated to explore.
> The most generous interpretation of Google's response is that Google's product suite is not aligned with the goals of educational institutions and that Google isn't willing to put in the work to get it there.
This was my read. Google has turned on a dime, comparatively. Their search model has transformed completely.
Consider that the whole web was their partner, and it worked well for decades. But ai has shifted the ecosystem and google is cannibalizing the web and their prior search experience to build a closed agentic search experience instead.
My read is schools and school revenue simply aren't that important to Google relative to their overall revenue and more importantly, overall profit.
And it DOES suck for schools that want to go no ai and who made investments, to be clear.
Schools in their current form will not be around. The Ai is 10/10 more knowledgeable than teachers. Maybe I’m being too broad but this 100% certain for the US.
> The Ai is 10/10 more knowledgeable than teachers. Maybe I’m being too broad but this 100% certain for the US.
I think this comment makes it pretty clear that this is just your personal experience and absolutely cannot be applied to every single teacher in the US.
Otoh, you have to be a super dumb pragmatic moron if you let Google or Microsoft into your school. If you hear someone takes offense with this truth, tell them to focus on the 'pragmatic' qualification; I heard only goods things about that word.
> under the assurance that these things will improve education
I'm 100% sure there is no contract that commits Google to "improving education". Even the existence of laptops in the classroom is or was highly controversial, because lots of people feared it wouldn't improve education. Google's responsibility under these contracts will be limited to supplying laptops and IT services, not to take over the educators entire job while letting them keep their salary.
> The most generous interpretation of Google's response is that Google's product suite is not aligned with the goals of educational institutions
The most generous interpretation of Google's response is that the school's requests make no sense, so Google is attempting to navigate a morass of emotional and ideologically driven hysterics for which there is no winning move.
To recap: A search engine generates pages of links to web sites. There is no point in having a search engine if you can't click links and browse the open web. Yet the internet is by now full of LLMs. Your local dishwasher repair center has an LLM chatbot on it which will, with a bit of coaxing and badgering, be happy to spit out whole essays on whatever topic you want. There is no such thing as "give me the internet without LLMs" at this point.
What are the schools expecting Google to do, exactly? Magically detect whether every website they link to has an LLM on it and silently delete those results? Stop students using the address bar?
It seems pretty clear that the schools demanding this are:
1. Ignorant of how widespread LLM tech actually is. They do a Google search, see Gemini answer and think that + chatgpt.com is the limit of where it's found.
2. Think something as trivial as hiding AI overviews on google.com will somehow stop students cheating, even though the kids have access to LLMs through basically every internet connected device they can find: on their phones, their tablets, their parents phones, on laptops at home, at internet cafés, through their TV, through their games consoles...
3. Advertising that they have exactly zero interest in figuring out the new world, even though that's what they're paid to do.
So they're asking a stupid question and Google is giving them a stupid answer. The schools need to up their game and figure out what post-AI education looks like, not throw tantrums where they demand the first tree they saw is chopped down while ignoring the forest.
> The request here is effectively that Google built and maintain a second version of their product with no front facing llm, and Google has declined to, correct?
When I happen to use Google search, the results page now has an 'AI summary' at the top, and a list of links below that. Could they just not display the summary and merely present the list of links (like they used to for the last ~20 years)?
I did a little research on this and it reminded me that even though Google generally stays out of the news (for bad things at least) they are still a giant corporation and are purely in it for profit, even if it's more lowkey. Forcing schools to have an AI feature enabled in ordered to use your product is not something I can agree with I'm afraid.
"Google made the surprise announcement two weeks ago, on Election Day, with executives saying they wanted to address teachers and parents’ concerns around AI in the classroom. Since then, communities across the country have protested Google’s tight hold on tech in many schools."
---------------
This policy has a "got to get them young" feel to it. Google seems to be more interested in hooking kids on their product than in letting educators do what's right for teaching. DuckDuckGo immediately offered an AI free version of their search as an alternative, putting the lie to any claims Google might make that it's difficult for them to do the same.
Google's executives are looking even more evil than usual on this file. They had better backtrack quickly.
Google has been indoctrinating kids for nearly two decades now. They've been essentially giving away Chromebooks for free in schools since Chromebooks were a thing
As someone with a teenager who has a chromebook issued by the school district I am very confused by this article. The IT department at my son’s school has clearly done something to his chromebook, when he searches Google an ai overview does not appear. Ever. There are no ai tools on his Chromebook. Is this a phased rollout over larger districts? The article does not provide enough information for me to make any assumptions especially when the core argument doesn’t resonate with my reality. Not only does his Chromebook not show Gemini the ethos (at least in my area) is that students don’t want to use AI. In fact, when reading anti ai sentiment on HN I’m often reminded of his discourse. That is not to say kids won’t try to to cheat; but while some try: when a 15 year old tries to submit homework written by an AI the teacher can spot it from a 100 mile distance. Getting these kids to read anything longer than a sentence is almost impossible. So many of the other kids in his class have exceptionally low literacy rates and emotional problems (which seem to be brought about by the combination of Covid + iPad parenting) that I would argue AI usage is a low ranking issue. In fact I would go further and say if the AI can write something they are willing to read maybe we should just let them. The world for them won’t look anything like the world we grew up in and while (if true) I would be very angry that Google is pulling their support for students I think this isn’t that and that there is a strong anti ai political aspect to this story which I believe is overstated
I cannot stand the Chromebooks. I wish they would kill them off. My kids got them as well but they at least still use text books. They are not using them at home though. They got steam decks attached to a screen for their gaming computers and its made the install mods and other stuff actually learning some computing.
Can someone help me understand why can't the schools install another browser? Or default to another search engine? I thought that this was possible on chromebooks. While I understand that this may be a second best solution, can't the IT staff at the school boards implement these solutions?
I used to work at a state college and during a major tech turnover we had 1 tech employee for 3 campuses across 3 counties. That 1 was me. 5000 students, hundreds of staff, and me keeping the lights on for 3 years.
The instant someone offered me $1 more an hour I quit.
K-12 IT employee here, I lurk HN a lot and made an account to throw in my 2 cents.
Anyway, both are generally true. My district uses a hub model, where support staff are dispatched out to schools and network infrastructure/business ops are centrally managed. But our model is common to larger districts; smaller ones often rely on small general-purpose IT teams - maybe a handful of people.
Obviously, being both public-sector and education, the pay isn't great because the funding isn't there. I like to think most of my colleagues are competent, service-minded people, we're just understaffed. We don't usually experience a lot of turnover, just growth from 2020-2024, but the economy has made some folks reluctantly move on lately. I'm hiring now and the strongest candidates withdrew when they realized what our salaries were. We're in a metropolitan area - rural districts have even weaker candidate pools and lower funding.
My district's leadership has embraced AI as part of our students' future, so they're taking an "allow it, but regulate it" approach and Google's change doesn't impact us. I have mixed feelings about how tech is integrated in grade schools in general, but this _is_ something of an added burden to workaround for under-resourced districts who don't want it.
Its not like there aren't other options. It seems unreasonable that a user who (presumably) isn't paying should be able to dictate product decisions. If you dont like google's decisions, vote with your feet.
The only reason Microsoft didn't get broken up over this behavior is the travesty of the 2000 election coming just before the judge was going to apply the penalties, and the Bush Jr administration putting the kibosh on it.
That then, of course, set the stage (well, alongside the Reagan-era shift in enforcement) for the subsequent 2+ decades of kissing the boots of oligarchs and acting like massive corporate consolidation is normal rather than an absurd outlier that flies in the face of a century of antitrust law and practice.
I don't Google for answers as much as Google seems to think I should. I often Google to navigate the web but they are making this actively harder. Plus their LLM answers are frequently either missing the point or flat out wrong. I feel bad for those that will grow up taking the generative AI answers as verbatim because they do not know better. Though there is some hope, some kids I've talked to have been taught to distrust LLMs and to ignore or not trust AI results - so at least we have that.
I haven't worked there in years but the visible heads of the AWU include some of the most clueless, dysfunctional people I have ever encountered and did not seem to me to represent anything resembling the central position of the typical Googler.
AWU is an unserious organization and absolutely does not have its act together. They seem ambivalent toward recruiting enough members to actually gain worker bargaining power (which like, isn't that the primary mechanism and purpose of a union?). Even before they have that power they want to charge you 1% of your salary to print t shirts and hold solidarity with contractors basically - the cart comes before the horse. This is literally just strongly worded letter without that power. Doesn't mean nothing will come of it (I hope it does much like the voluntary exit program), but if it is resolved, it will likely be due more to the larger outcry and only in very small part to AWU itself (again like the voluntary exit program). Very expensive socialist social club otherwise.
And which would you propose? There's no search engine I know of that isn't run by some kind of profit-seeking corporation. And that's probably a good thing, since we all know what public sector funding for libraries ends up looking like.
In related news, Google punishes several employees for insubordination. Didn't they learn their lesson when this exact same thing happened with the Palestine cause?
It seems like a pretty simple idea for schools to just switch to ddg or bing if they choose to keep their students ignorant of AI until it replaces them.
Plus I don't get the point of petitioning a business to do something, it ain't a charity.
1. People petition businesses all the time, especially when the business has a monopoly. Besides regulations, things like the MPAA,ESRB, and even just Xbox backing down on some anti-consumer changes all are the result or (consequence if involuntary) of this.
2. Google absolutely behaving nominally charitably around schools. The strategy around things like Chromebooks may have a well troded ulterior motive but that isn't on the books.
1. What does google even have a monopoly on? Search? I just told you about ddg and Bing. About workplace suites you gotta be kidding me xD. Cuz there O365, Zoho, even proton has one.
2. When a business behaves charitably you just know they have some ulterior motives. I wouldn't trust them.
It's frustrating because you have to commit a non-negligible amount of your paycheck to the union and then it seems like this particular union (AWU) spends most of their effort helping non google employees (vendors, extended workforce, contract workers etc) negotiate what they want. I think if they thinned out their expenses to require lower fees and focused on larger issues it would be much more effective and have larger bargaining power.
As someone who spent a fair amount of time at Google trying to get people to join the union, this is 100% true. Almost everyone who I asked about it cited union dues as their main concern. Extremely frustrating.
Perhaps it would be easier for the union to get people to join if it did things that would actually benefit them instead of basically being a PAC for the omnicause, supporting things that directly oppose employee interests.
When I left Google like seven months ago, the union had decided (based on feedback from members) that it was primarily interested in job security related things, such as buyouts and severance packages. I don’t know if that’s changed or what.
Well, it certainly wasn't their concern for its first five years of existence, and its external communication only barely mentions job security for its members: https://www.instagram.com/alphabetworkers/
They need 50% + 1 percent of the bargaining unit. Including others (vendors, extended workforce, contractors) in that not only makes it so that you need more people, but also that those individuals have conflicting goals which the union cannot represent (FTEs may have a workplace issue that conflicts with vendors).
As such, it cannot be a union registered with the NRLB and so will never have the bargaining power.
If it was instead a single group (e.g. subcontractors working for Google Fiber) then it would be easier to get the 50% vote and would have a single focus and could be a union that has collective bargaining power.
Well I guess 1% of pretax income is non-negligible depending on what you get out of it. If its a worthy cause sure, I probably donate closer to 2% every year to various things.
When it's a charitable donation to fund labor lawyers and organizers for employees of Accenture and Cognizant, its a no for me.
I mean, my favorite charities would be happy to accept 1% of your salary.
1% of salary seems like a significant amount. Either you're not making much and every dollar counts (but then you're probably not a direct Google employee), or you're making enough that 1% is real money that you could do something with.
SAG-AFTRA charges 1.575% on the first $1M of income plus a base of about $250, but it's an industry union with significant market power; membership comes with benefits to weigh against the dues and other drawbacks.
What benefits is AWU providing for members? How do they weigh against the dues and other drawbacks?
lol, the AWU is a joke. As a sibling comment pointed out, it's a PAC disguised as a union.
A quick look at the demographics of their members page and comparing it with the demographics of the average Google employee will tell you why.
As a minority union, AWU can’t force Google management to come to the table to negotiate. But it can try to pressure executives by rallying members toward a cause. That’s exactly what happened last week, when AWU called on YouTube to permanently ban Trump in the wake of a raid on the US Capitol on January 6th.
“YouTube refuses to hold Donald Trump accountable to the platform’s own rules by choosing only to remove one video instead of removing him from the platform entirely,” they wrote. The company “is avoiding the proactive action called for by both their workers and the broader public.”
ON JANUARY 4, a group of workers at Google parent company Alphabet said they had formed a union to influence the company’s approach to political and social issues. Three days later, the Alphabet Workers Union got its first chance. A crowd incited by President Trump had stormed the Capitol in Washington, DC, sparking recriminations in the capital and in the atomized home offices that now make up Silicon Valley. YouTube had deleted a video the president posted amid the violence on Wednesday. But unlike Facebook, Twitter, Snapchat, and Amazon-owned Twitch, YouTube did not suspend Trump’s account.
The Alphabet Workers Union said YouTube’s response was “lackluster” and inadequate. In a statement, the group said Google was not enforcing its own policies, while the platform was being used to “spread hatred and extremism.” “YouTube will continue to function as a vector for the growth of fascist movements if it persists in prioritizing advertisers while exposing the public,” AWU wrote. Google did not respond to requests for comment, but told Fortune that the president’s account had not violated its three-strikes policy.
The AWU is somewhat unusual in that it is affiliated with the Communications Workers of America but won’t seek recognition or collective bargaining rights through the National Labor Relations Board. “We will use our reclaimed power to control what we work on and how it is used,” AWU writes in its mission statement. “We will ensure Alphabet acts ethically and in the best interests of society and the environment.”
----
It wasn't focused around workplace issues and instead was trying to force YouTube to deplatform Trump and have Google act in certain ways including the "control what we work on and how it is used" part.
AWU may have good and righteous issues... and some were workplace focused (termination and harassment)... but its largely been politically focused in its inception and messaging.
Maybe unpopular opinion, but unions should focus on improving working conditions, wages, and protecting workers from retaliation or unfair dealing not trying to influence the direction of the business. If a union were to push for the issue in TFA and it turns out to be the wrong decision, who is responsible? Will the workers take a voluntary pay cut to compensate for lost revenue? It also gives the executives another scapegoat if things do blow up.
>unions should focus on improving working conditions, wages, and protecting workers from retaliation or unfair dealing not trying to influence the direction of the business.
There is precedent for this in Germany, at least, though I don't know anything other than it exists.
In Japan (maybe other nations), unions can have a seat on the Board. They advocate for the employees, but as stakeholders in the enterprise, which I think is smart.
Not everybody knows this but you can actually have collective bargaining without the world ending.
I think the key difference for countries outside the US is they have stronger national cohesion and stronger national identity. The US is a country of immigrants (always has been, always will be, literally) and so it's harder to feel solidarity.
The US has an extremely strong national identity. The problem the US has is that everyone considers themselves to be a temporarily embarrassed millionaire.
The exploited underclass don’t see themselves as the exploited underclass, and so they have no desire to band together to improve their situation. They believe that they too will become rich if they continue to play the game they’ve been told to play.
"Maybe unpopular opinion, but unions should focus on improving working conditions, wages, and protecting workers from retaliation or unfair dealing not trying to influence the direction of the business."
That is the bulk of what union workers are doing every day through their unions.
That doesn't mean unions shouldn't try to intervene on issues of broader concern as well. It's not a zero-sum game.
> If a union were to push for the issue in TFA and it turns out to be the wrong decision, who is responsible? Will the workers take a voluntary pay cut to compensate for lost revenue?
If the executives make the wrong choice, it's already the workers who will make up for it with an involuntary pay cut or layoffs.
Google _has_ a union, as others have pointed out. Unfortunately they're too busy virtue signalling and tilting at windmills i.e., doing everything but looking after employees' needs, which, AIUI, is the main purpose of an employee union.
Layoffs, forced "VEP"s, etc. are happening all over Google, but the union is more concerned about Gaza, Palestine, etc.
i don’t understand, what is Google if not their employees?
i know i know, I’m being deliberately obtuse, but I just hate this notion that people are just employees and they throw their hands in the air or shift blame to management.
if you are an employee you are responsible to the eyes of the people
This seems like a reasonable stance and not an impossible one to defend. Google is making a bet that long term, everyone will use AI. I think that's undeniably true.
The friction here is that they don't want to bifurcate their main offers to maintain an "AI free" experience, because the thesis is this won't exist long term. I think this makes sense.
This is not opposed to telling employees not to "punish schools", whatever that means. Nobody is being punished, they're just not being catered to.
No, even if the schools enable AI, it doesn't always show up on search. But it's a bet on people using it far more than they do today, which seems right. We're already past the tipping point where frontier models are more helpful than search engines for information retrieval, and soon that will be distilled down into ultra cheap models and nobody except Luddites will keep insisting on doing searches manually.
I think the Luddites have a case because LLMs are irrelevant to the "search is your address bar" use case. Google took off when it became easier and faster than remembering the URL when you had a specific page in mind. The AI-summary contributes nothing to this except moving the link you want to click on several centimetres down the screen.
I’ll type a one word search term to find a site, the result I want will be at the top, and Gemini will pump out a completely irrelevant half-page summary about something completely unrelated to what I was looking for.
Maybe don't frame everyone whose not a FOMO AI booster as a Luddite? Some of us want balance and nuance, and to treat AI as normal technology.
It's mildly insulting and if you know your history, inaccurate.
Personally, I'm hoping that once the FOMO AI boosters throw around the term 'Luddite' enough it'll become a positive term praising people who carefully evaluate risks/benefits of new technology instead of thoughtlessly chasing the FOMO.
Luddites had a good political position, and their demands were reasonable. Of course, the capitalist classes used the power of the state to put them down with violence, because police exist to protect the property of the wealthy, not to actually support justice.
No, Google has made a bet that forcing people to see AI search results will somehow be good for them. The moment the AI search results appeared, people asked for a way to turn it off. The answer was no.
They're not the only company refusing to let people turn off AI features people don't want. In this case, it doesn't appear to matter what the customers want.
I don't really see a problem here. Google has the right to design their product how they want to, and users have the right to stop using it if they don't like the product anymore.
The claim that "customers don't want this" isn't backed up by data. What percent of customers don't want it? What percent of customers actually do want these features and use them all the time?
That would have been nice before they obtained the entire market, but I guess school districts can just switch and retrain over the weekend? Schools have tons of money too so they can dump all those chrome books and buy iPads.
My assumption of clarity was, apparently, unfounded.
Google has what would have been regulated away as a monopoly when there was an appetite for that kind of fight. It's an analogy, not a metaphor.
There are so few competitors because Google has systematically starved them or bought them, so that it is practically impossible for anyone to compete with them on product or reach. It is not because they are the only tech company capable of building a good search experience.
Google Search has 5 billion monthly users. I would be surprised if anti-AI users were even close to 50M, i.e. it's likely far less than 1% of the userbase. But if you want to make that argument, I would be interested to hear it.
I cede that some percent of Google's users are anti-AI, because that's obvious. What I want is for the person making a claim to be specific about what they're claiming. Are we talking about 50% of users, 1%, or 0.001%?
You do realize that Google would be the only organization capable of getting that data definitively right? And they have zero incentive to ask the question?
A major municipal entity in a major municipality, we're talking about the entire population of their remit when we discuss customer impact. But focusing on something like 'percentage of total users' ignores the critical 'user value' dimension.
Actually, they do have the right to do that. Companies do it all the time, that's how we get better products. Or they run the product into the ground and the users move on to better things. That's how the process of innovation through competition works.
I have to disagree on the punishing aspect. They enabled Gemini for student accounts and there was a significant backlash. They then added some configuration so that schools can disable AI features, stating that if all AI features are disabled then they will _also_ disable Google search.
If they didn't want it to sound punitive, another solution might have been more appropriate. Google search today has a "Web" item (listed under "More" at the top of the page) which does provide search reasults without the AI overview. I think this indicates that this feature isn't so difficult to provide to schools, their paying customers.
Even that would be significantly more defensible for the manufacturer. They don't make it anymore, that sucks but gotta move on.
It's more like the bottom half of the jar is the same amount of peanut butter as ever and then they add a big squirt of mayo on top. They still make the old product as part of the bundle, and the bundle is the same price, but they insist on only having the bundle.
Which is perfectly sensible, both for the schools (get peanut butter from somewhere else) and the manufacturer (believes mayonnaise is more profitable than peanut butter). I don’t know why the manufacturer’s employees need to lobby the manufacturer to sell a product the manufacturer doesn’t want to sell.
But google has always been an AI overview. Google filters out things you think don't matter - that's AI. Google judges what pages are useful and which ones are trolls - that's AI. Google analyzes your search and decides how to interpret the terms to bring you the most "relevant" pages in the context you're asking from - that's AI.
Google is AI to the very core, to the very marrow of its bones, and it always has been. It wasn't always llm, but to pretend Google isn't AI is just ignoring the statue and staring at the fig leaf and thinking that's the product.
You know what "the AI overview div" means. Arguing about whether the rest of google counts as "AI overview" if you squint hard enough is a distraction at best. The div, the one at the top, the one with an LLM and a reply box, that's the div.
There are many things called "AI". Smart search ranking driven by machine learning that helps me find relevant original content written by real people is one. LLM-generated slop is another. More of the first, less of the second, please.
This framing confuses me. Google has shifted its product to be a search engine with AI first. The request here is effectively that Google built and maintain a second version of their product with no front facing llm, and Google has declined to, correct?
It isn't like there is some second version of Google that Google is withholding from schools. However, there are search engines which don't use AI. Switching to those would probably do more than protests to get Google to switch its product roadmap.
are you someone from microsoft in the 1990's and trying to tell us microsoft explorer is integral to Windows?
Are you someone from Apple in 2008 and trying to tell me Safari is integral to Mac OS?
Not Safari & macOS! Next they'll come for the iOS!
Likely
Couldn't they just turn off the generative summary and go back to showing search results instead? It doesn't seem like asking for too much additional from one of the world's largest and most capable companies.
Technically, perhaps, but the future is in LLM answers. The old style search is likely to rot as Google invests everything in the AI mode.
Or another perspective: Nobody expects to be able to ask Google for a version without ads, even though ads are annoying, because they know Google makes most of its revenue from search ads. Why do they then presume to dictate that Google give up on their strategic priority of pivoting to AI? It's equally absurd. Thankfully we have long had a solution for this: client-side blocking.
Search has already rotted for years. “SEO” killed it years ago.
Oh I agree. But until recently Google engineers at least actively were fighting to save it. Now it's on its way to the graveyard.
I never understood this near-ubiquitous argument.
SEO didn't force Google to ignore half your search terms or give you results for thematically or semantically similar search terms that are ultimately irrelevant to what you actually searched for.
That was all the doing of employees at Google. SEO did not have even the tiniest iota of relevance for Google's discretionary choice to make search results worse.
But even though they are worse, they are not yet useless.
But aren’t these chrome books supplied to schools by google? So would they allow installation of client side blockers?
But thats my assumption, so I might well be wrong.
Chrome extensions can be used to suppress elements on pages, and schools can force install and enable extensions.
Only if chrome supports the required extensions, which with a drop of manifest v2 they only do half assed. School boards can also filter at the network level but that's very hard to get right and can cause a whole new set of problems
Ads pay for the product, and AI doesn't. It's absurd to ask them to burn money, it's not absurd to want a toggle for a new feature that causes problems.
And they'll fight back against blocking.
(Also asking for a version without ads is reasonable if you're willing to pay a modest fee for it. So that's not absurd either.)
It's not just any old new feature. It's the one they are betting the company on. "Google is an AI-first company", and all that.
That seems like it's Google's problem.
It doesn't get to make it ours.
Sure they do. Google is under no obligation to bend over backwards to make users of its (free) product all happy. It's a competitive market, if you don't like what Google offers, don't use it. What Google offered schools is strictly more control than before, and they complained. Lesson: offer less control?
The schools could already block google. This setting isn't more control, it's mockery.
And if you don't like a product change, maybe don't use it but also complain to apply pressure.
I'm sorry, what?
You think Google is in a competitive market??
They're exercising their market dominance to do this. If you can't see this, you're willfully blind.
So they could easily and they just don't want to?
No, Google still has a search engine that returns pages of results like it is 1999. The Gemini iframe (or whatever) sits on top of those pages of results.
Doesn't Gemini also use classical search as a tool?
It does not. The ranking is AI based, and has been for 12-15 years.
The idea of a pure Page-Rank ™ Google search is long gone.
The schools are very obviously against the AI Overview, not the unseen ranking technology.
Schools obviously don’t have an issue with using AI to rank search results, though.
They want to disable student access to the large language model Gemini.
This is just willfully misunderstanding what the schools want. It's very obvious they want the summary/chatbot interface to go away, and to only show links that are shown below the chatbox instead. Nobody is asking anyone to change the ranking.
This is not a particularly big ask. If it truly is lost science to not add a chatbox above the links, they could hide it it via the CSS layer if needs be. If nobody remains at google that knows how to hide a div with CSS, half the people in this comment thread could help them with that.
Their requirements are broader than the user interface. Some boards have policies to prevent students from using AI, or using student content for AI training. In other words, there can be no read or write pathways that include AI.
There is no reasonable way for Google to do that (that they could affordably sell to the schools).
So that leaves no obstacle for allowing schools to hide the chat interfaces and summaries! Great!
the chat interfaces are not the only pathway into AI on Google search.
Right, but it's the pathways most schools want to hide. So why not let them?
GOTO A:
What other interfaces are there? There's an AI tab at the top, there's the summary, and otherwise I don't see any.
It's embedded into individual result blocks, too. I've seen it in local place and stock ticker results.
The AI that may have proved the Riemann Hypothesis has finally met its true match: conditionally displaying web components.
I think you are projecting your own desires onto the schools, and if that were actually what the schools wanted, it would not be an issue
>> This is just willfully misunderstanding what the schools want. It's very obvious they want the summary/chatbot interface to go away, and to only show links that are shown below the chatbox instead. Nobody is asking anyone to change the ranking.
> I think you are projecting your own desires onto the schools, and if that were actually what the schools wanted, it would not be an issue
It's pretty obvious you are wrong and the GP is correct.
If Google's ranking was done by AI for the last "12-15 years", the schools didn't care. But now Google adds a fucking chatbot that the kids can ask to do their homework, and that's when the schools start caring. It's pretty obvious the chatbot and AI overviews are what they care about.
Certainly it's a fact that the schools didn't care in the past, but I think that's mostly a result of AI not being in the spotlight then, and there wasn't a broad anti-AI faction then.
From TFA:
> Attendees of all ages held signs reading, “NO AI IN MY SCHOOL!”, “A.I. Gets an F,” “Stop stealing our imaginations,” and “SCHOOL Librarians = THE BEST Search Engines.”
I'm pretty sure these people would not be okay with plain search if they understood the extent to which it uses and has used AI.
(Also, I suspect you and other commenters have an implicit belief that the AI plain search uses and what the AI box does are fundamentally different, and I doubt that. The search algo is going to incorporate new tech as it comes along, and it's unlikely there is zero LLM involvement.)
> Certainly it's a fact that the schools didn't care in the past, but I think that's mostly a result of AI not being in the spotlight then, and there wasn't a broad anti-AI faction then.
> From TFA:
>> Attendees of all ages held signs reading, “NO AI IN MY SCHOOL!”, “A.I. Gets an F,” “Stop stealing our imaginations,” and “SCHOOL Librarians = THE BEST Search Engines.”
> I'm pretty sure these people would not be okay with plain search if they understood the extent to which it uses and has used AI.
I doubt it, and if they did, it'd probably be due to misunderstanding where they conflate LLMs with pre-LLM AI technology (e.g. the "12-15 years" tech that was used for Google Search ranking). The emotions are pretty clearly being driven by the current thing (LLMs and similar models).
I think you're making the mistake of interpreting their statements too literally, using definitions and interpretations that they simply don't share, to falsely twist their statements into embarrassing contradictions. You can use the word "AI" and they can use the word "AI" and you can both mean different things (or have some conceptual overlap but with significant differences).
> (Also, I suspect you and other commenters have an implicit belief that the AI plain search uses and what the AI box does are fundamentally different, and I doubt that. The search algo is going to incorporate new tech as it comes along, and it's unlikely there is zero LLM involvement.)
When they say AI, they are almost certainly referring to the chatbots, the AI overviews, etc. and excluding search, which has no user-obvious AI component. Being "but ackchyually" to try to turn their position on its head can be pretty disrespectful.
School board's set policy at a high level. If they pass a resolution "no student facing AI" -- now Google has to turn a lot of things off that were on before.
What do people gain from being obtuse? You want to pretend to be confused even though you're probably smart enough not to be? It's very simple, people want the traditional list of links search results and not LLM-written overviews. It's not like that has been thrown away or isn't still in production.
It's a correction to the motherjones article which (sadly, like 90% of their articles) is stated as disingenuously as possible.
They seem to be saying that Google is taking some punitive action directed at schools that don't adopt AI.
Whereas Google is actually just making an AI-only search product.
The move is 100% punitive on the part of Google. From the recent CNN article:
"Google is rolling out new controls for schools to limit students’ access to the company’s AI tools. But if schools opt out of AI altogether, they’ll also lose access to Google Search."
They are saying that if schools would like to disable all of the AI tooling in the Chromebooks used by students then Google will also disable access to search.
>They are saying that if schools would like to disable all of the AI tooling in the Chromebooks used by students then Google will also disable access to search.
THAT is a clearer answer than any I received. If the demand is "let us turn off AI features generally, but keep search with AI" then that seems like an extremely reasonable request. I'm not sure if that is the request but it's not outlandish vs. maintaining a contractually provisioned no AI search engine version.
No?
They are very clearly saying they want the AI summary and chatbot in the search results gone.
Because it is a way for students to cheat.
No.
What they are aiming towards is irrelevant. Your argument could just as well be interpreted as "Google is trying to mandate students rely on AI summaries so Google can one day stop linking to websites that are not owned by Google".
If anything, this is an argument in favor of the schools and against Google.
If Google wasn't still linking to external websites then you might have a point. But they are. So what Google wants to "nudge" students to do is irrelevant when schools - the entity actually empowered to teach students - are asking them to hide an iframe when students are using search.
Until the day Google stops linking to the rest of the Internet, that will be the most trivial of asks.
> The request here is effectively that Google built and maintain a second version of their product with no front facing llm, and Google has declined to, correct?
Google already has a web search with no front-facing LLM, and you can get it by selecting "Web" from a search results page, or by including udm=web in a constructed search URL like so:
https://www.google.com/search?q=deliberately+obtuse&udm=web
Somehow, it's impossible for Google to provide schools with the no-AI-summary search engine they already provide! My personal opinion of Google's behavior is in the example URL that I gave.
Or by including -ai at the end of your search query
Or by including -whatever at the end of your search query... It's not a specific "-ai" parameter, AI isn't yet trained to answer negative queries so if Google sees a negative query it disables AI overview.
The ordering of the search results differs depending on whether you include "udm=web".
Compare:
https://www.google.com/search?q=deliberately+obtuse
https://www.google.com/search?q=deliberately+obtuse&udm=web
This is interesting.
hm.. the order doesn't differ for me, but `udm=web` shows 10 results, and without it I only get 8
Are the schools explicitly asking for "no front-facing LLM"/"no AI-summary", or are they asking for "no AI"? Those are two very different things. Separate from the LLM-generated summaries, the basic Google Search results ranking have used various AI techniques for well over a decade. It should be easy for Google to turn off the summaries for specific organizations, but it's infeasible for them to go back to the pure Pagerank results from the 2000s.
https://www.nytimes.com/2026/11/18/technology/google-gemini-...
> Starting next year, Google said, its A.I.-related products, including Search and its Gemini app, will be turned off automatically for schools, and administrators can choose to turn those A.I. tools on. But there will be no version of search available without A.I. chatbot features, meaning schools that want to forgo A.I. will lose access to an important learning tool.
It's specifically chatbots that Google is saying they won't allow schools to disable without disabling search as a whole.
> Are the schools explicitly asking for "no front-facing LLM"/"no AI-summary", or are they asking for "no AI"? Those are two very different things. Separate from the LLM-generated summaries, the basic Google Search results ranking have used various AI techniques for well over a decade
The first one. They've made that clear, and also, evidence over the past decade+ indicates they aren't concerned with the second one. And why would they be?
> It isn't like there is some second version of Google that Google is withholding from schools.
This is literally exactly what is happening.
There is no second version. It hasn’t existed for 12+ years
https://www.google.com/search?q=There+is+no+second+version.+...
Ex googler says there are actually millions of versions: https://news.ycombinator.com/item?id=50581469
abiding by a "no AI" policy means not presenting or contributing to AI content. AI is not limited to AI chatbots (like chatgpt). it would mean any read or write path that leads to AI training or inference.
Google search content reads and writes AI pathways, even on "normal" search (aka Web search tab)
Do you have evidence of that? Because personally i would be very surprised that google is maintaining a non-AI version just for kicks and not using it anywhere. That is just not how tech companies operate.
https://www.google.com/search?q=confidently+incorrect&udm=we...
Search any NSFW term. Google will give you a non-AI results page.
It is exactly how Google operates, they have a high-level classifier that runs on every search query to determine which widgets show up on the page.
Or search "who won the 2020 election". No AI overview!
They have to maintain a web search engine regardless, because the LLM needs it. It's not just hallucinating facts from its weights. (I mean, occasionally it will do that, but usually it's based on search results.)
This isn't about maintaining the back end. It's about how the search results are presented.
> It's not just hallucinating facts from its weights. (I mean, occasionally it will do that
My experience is the Google AI block at the top of each search has very frequent hallucinations presumably caused by the fast model required so users don't scroll past it before it's finished loading.
> Google has shifted its product to be a search engine with AI first. The request here is effectively that Google built and maintain a second version of their product with no front facing llm
Guys, we're all programmers here. We all know this is the kind of bullshit engineers say when they don't want to do something. We don't have to fool each other.
I don't know how Google Search works, but I can guarantee you that they don't have LLM responses intrinsically linked to search results. If their LLM services go down, I'm pretty sure they can still serve search results. All that schools want is to have that flag exposed in the account/organization settings page.
This is the same type of bullshit Microslop tried to pull with Windows XP: saying that they can't unhook Internet Explorer from Windows. The EU forced them and it turned out it was possible all of a sudden.
Saying "maintain a second version" is so hyperbolic. All that's being asked for is making a flag public. A single Google engineer should be able to code it (even without AI) and deploy it in 2 hours worldwide.
Yes &
> we're all programmers here
Although I’m not I’ve been served the B of the A/B test: no summary at the top. (I avoid Google a bit but this is as recently as ~last month.)
Every widget on Google’s search page is an independent component maintained by different teams. There are thousands. Each one can be configured to trigger or not based on thousands of different switches (environment, product config, experiments, etc.)
It’s dead simple to turn off singular components in specific configs.
Source: Worked on a search team
"web" seems to have most of that turned off, so this is confusing in that the requirements are mostly there already.
Right now, but once you contractually sell a "no AI" product then you need to maintain a version which has "no AI". The existing of a link form which has no AI presently is very different from a contractual commitment to institutional buyers who will complain loudly and sue if the contractual conditions are not met.
No you just let someone pay to be on an IP address list and when handling a request, if the client IP is on that list, you skip over the section of the code that generates the AI summary.
What about a student logged in from home?
What about them?
They are on a different IP address. Why would you tie choices to the address rather than the account?
Sure, you could do both.
And if they get around the restriction or find a way to ai mode is Google contractually liable?
To feel confidant in the delivery I would expect Google to need to build and maintain a no-ai search engine, indefinitely, even if their current search experience becomes more ai-centric.
This was my point. I think Google knows their own project roadmap and doesn't want to contractually build and maintain a legacy no-ai version of search.
Well there’s already &udm=14 and options to disable search buried in google account settings.
But I guess they might stop selling search and just sell ai?
I also work in Search and no I don't think launching this is "dead simple".
You're trying to make us believe that it's nigh impossible to add '&udm=14' to a search query?
Google sold this product to many schools during the pandemic. In my opinion, it is an epic fail on their part that they have ignored the needs of this market segment. That they now expect them to come up with their own solution to a problem that Google has created seems unreasonable to me.
This is a trivial change for Google. I worry that the rationale is that if they force-feed the AI summaries to children then those children are more likely to grow up into customers willing to purchase LLM products or be generally more willing to accept LLM content. Which is kind of gross.
Won't someone think of the poor multi-trillion dollar company having to make a simple feature flag available to organizations
Based on the article they linked, it doesn’t appear to just be protests:
https://www.nytimes.com/2026/11/18/technology/google-gemini-...
Although “discussing alternatives” is an incredibly modest step away from Google.
This is the dumbest strawman argument, and I can't believe people are still stooping to it to carry water for Google.
Schools do not care whether Google uses AI under the hood. They do not care whether Google sorts its results using AI.
They do want Gemini to stop answering student questions when they're typed into Google, and to stop giving summaries at the top.
The dumbest, working solution would be a one-line CSS fix to hide the Gemini box at the top of search results for schools. A smarter solution is easily within the grasp of Google.
Exactly. If the teacher asks the student a question and the computer does it for them, then what's the point? Your homework is to type this into Google and copy and paste what comes out.
Their homework could also turn out to be learning "why not to trust the first thing AI tells you", another valuable lesson.
You must be very easily confused then. The version without AI already exists
Reminds me of a certain TV personality and troll with the perpetually confused face.
https://www.google.com/search?q=confidently+incorrect&udm=we...
If customers are literally asking for a certain customization like this, I don't see why it's a crazy ask. I don't hate the AI features in Google but I think it makes sense to have a setting somewhere to disable it.
The request is to make the visible bits (AI mode, AI overview) hidden. Schools don’t care if Google is using classical PageRank or AI-based ranking under the hood.
Here is the part you are missing: Google sold Chromebooks and G Suite to schools for the last 15 years under the assurance that these things will improve education. Schools are now telling Google that Gemini is a serious problem for cheating and coursework, and Google's response is that they can't be bothered to put in a feature flag for this new feature that is causing problems. For institutions that have collectively spent tens of billions on locked down Google hardware, that is a thumb to the eye.
The most generous interpretation of Google's response is that Google's product suite is not aligned with the goals of educational institutions and that Google isn't willing to put in the work to get it there. The actual story of course is that Google is a monopoly that lobbied hard to get schools locked into their ecosystem and now they are pushing their weight around to try to get kids hooked on their AI early.
Artificial intelligence has turbocharged my ability to learn things.
But I'm CHOOSING what I want to learn. Or learning what's required to solve real world problems that matter to my success.
Schools are DICTATING what students learn, more often than not. And children will escape being controlled when possible.
The educational model is broken. It was broken when it was invented.
Real learning is very different. It's the most powerful thing humans do and our system stifles, bores and turns off a lot of children.
Most spend 12 years in school and come out with no marketable skills. Something could be wrong with that.
What's something you are learning now and how? I'm curious. What's something you'd recommend everyone learn?
> Most spend 12 years in school and come out with no marketable skills. Something could be wrong with that.
You’re not wrong, but schools are only partly to blame for that. Plenty of other factors out there. Don’t make the mistake of blaming schooling for a broken economic system and what are fundamentally social class issues.
They are wrong. 17 year olds don't need marketable skills. They need experience, and an opportunity to build on their their substrate of general knowledge and deeply study one or two areas that interest them.
If school isn't delivering that, then yes, they are broken. But no-where in schooling is the requirement to deliver someone with "marketable skills".
Aside from mathematics, which I use professionally on a daily basis, the most impactful lessons I learned in high school were in home economics. I’m not sure that’s not even taught anymore, which is a shame. I learned how to run a kitchen and keep it clean, how to plan a personal budget, how to do basic clothing repair, sewing, and so on. It all sounds mundane but I grew up in a house where some of those things (cleaning in particular) did not happen and it was a real revelation to me. I don’t think 14-year-old-me would have chosen to take those classes, but I am so glad that I did.
> Most spend 12 years in school and come out with no marketable skills. Something could be wrong with that.
There's nothing wrong with that. That's the base for learning "marketable skills". 12 years gives you a solid grounding in basics, and you start to get an idea of what you're maybe good at, or interested in, and then you get the opportunity to diversify, to take that basic learning to university or college and study one or two specific things, or get some kind of entry-level/junior position.
17 year olds don't need "marketable skills". They need encouragement, opportunity, and experience.
And anyway, fuck your "marketable skills". Knitting is not a marketable skill, but you know what, it shows skill, application, and dedication. Not everything learned in school is marketable, and there's nothing wrong with that and everything wrong with thinking otherwise.
Foundational education shouldn't be up for debate. It should be curriculums designed by experts to give us a foundation of knowledge to make responsible choices about what comes next. It's why you don't get to pick classes till high school, and certain types of courses are still required in college. Sometimes you need to know stuff you aren't motivated to explore.
> The most generous interpretation of Google's response is that Google's product suite is not aligned with the goals of educational institutions and that Google isn't willing to put in the work to get it there.
This was my read. Google has turned on a dime, comparatively. Their search model has transformed completely.
Consider that the whole web was their partner, and it worked well for decades. But ai has shifted the ecosystem and google is cannibalizing the web and their prior search experience to build a closed agentic search experience instead.
My read is schools and school revenue simply aren't that important to Google relative to their overall revenue and more importantly, overall profit.
And it DOES suck for schools that want to go no ai and who made investments, to be clear.
Schools in their current form will not be around. The Ai is 10/10 more knowledgeable than teachers. Maybe I’m being too broad but this 100% certain for the US.
> The Ai is 10/10 more knowledgeable than teachers. Maybe I’m being too broad but this 100% certain for the US.
I think this comment makes it pretty clear that this is just your personal experience and absolutely cannot be applied to every single teacher in the US.
Plus: School isn't only about knowledge.
Otoh, you have to be a super dumb pragmatic moron if you let Google or Microsoft into your school. If you hear someone takes offense with this truth, tell them to focus on the 'pragmatic' qualification; I heard only goods things about that word.
> under the assurance that these things will improve education
I'm 100% sure there is no contract that commits Google to "improving education". Even the existence of laptops in the classroom is or was highly controversial, because lots of people feared it wouldn't improve education. Google's responsibility under these contracts will be limited to supplying laptops and IT services, not to take over the educators entire job while letting them keep their salary.
> The most generous interpretation of Google's response is that Google's product suite is not aligned with the goals of educational institutions
The most generous interpretation of Google's response is that the school's requests make no sense, so Google is attempting to navigate a morass of emotional and ideologically driven hysterics for which there is no winning move.
To recap: A search engine generates pages of links to web sites. There is no point in having a search engine if you can't click links and browse the open web. Yet the internet is by now full of LLMs. Your local dishwasher repair center has an LLM chatbot on it which will, with a bit of coaxing and badgering, be happy to spit out whole essays on whatever topic you want. There is no such thing as "give me the internet without LLMs" at this point.
What are the schools expecting Google to do, exactly? Magically detect whether every website they link to has an LLM on it and silently delete those results? Stop students using the address bar?
It seems pretty clear that the schools demanding this are:
1. Ignorant of how widespread LLM tech actually is. They do a Google search, see Gemini answer and think that + chatgpt.com is the limit of where it's found.
2. Think something as trivial as hiding AI overviews on google.com will somehow stop students cheating, even though the kids have access to LLMs through basically every internet connected device they can find: on their phones, their tablets, their parents phones, on laptops at home, at internet cafés, through their TV, through their games consoles...
3. Advertising that they have exactly zero interest in figuring out the new world, even though that's what they're paid to do.
So they're asking a stupid question and Google is giving them a stupid answer. The schools need to up their game and figure out what post-AI education looks like, not throw tantrums where they demand the first tree they saw is chopped down while ignoring the forest.
> The request here is effectively that Google built and maintain a second version of their product with no front facing llm
I only have a cursory familiarity of this story, but isn't https://www.google.com/search?q=foo&udm=web already a "version of their product with no front facing llm"?
> The request here is effectively that Google built and maintain a second version of their product with no front facing llm, and Google has declined to, correct?
When I happen to use Google search, the results page now has an 'AI summary' at the top, and a list of links below that. Could they just not display the summary and merely present the list of links (like they used to for the last ~20 years)?
Google actually does have a version of its search engine without AI.
If you search something and then click the “Web” option at the top then it just returns the 10 blue links.
The option is next to where the options for “News” “Images” “Books” etc are.
I did a little research on this and it reminded me that even though Google generally stays out of the news (for bad things at least) they are still a giant corporation and are purely in it for profit, even if it's more lowkey. Forcing schools to have an AI feature enabled in ordered to use your product is not something I can agree with I'm afraid.
Nominally, yes. In practice, employees get promoted based on their token burn rates so stuff like LLM "search" happens.
"Google made the surprise announcement two weeks ago, on Election Day, with executives saying they wanted to address teachers and parents’ concerns around AI in the classroom. Since then, communities across the country have protested Google’s tight hold on tech in many schools."
---------------
This policy has a "got to get them young" feel to it. Google seems to be more interested in hooking kids on their product than in letting educators do what's right for teaching. DuckDuckGo immediately offered an AI free version of their search as an alternative, putting the lie to any claims Google might make that it's difficult for them to do the same.
Google's executives are looking even more evil than usual on this file. They had better backtrack quickly.
Google has been indoctrinating kids for nearly two decades now. They've been essentially giving away Chromebooks for free in schools since Chromebooks were a thing
As someone with a teenager who has a chromebook issued by the school district I am very confused by this article. The IT department at my son’s school has clearly done something to his chromebook, when he searches Google an ai overview does not appear. Ever. There are no ai tools on his Chromebook. Is this a phased rollout over larger districts? The article does not provide enough information for me to make any assumptions especially when the core argument doesn’t resonate with my reality. Not only does his Chromebook not show Gemini the ethos (at least in my area) is that students don’t want to use AI. In fact, when reading anti ai sentiment on HN I’m often reminded of his discourse. That is not to say kids won’t try to to cheat; but while some try: when a 15 year old tries to submit homework written by an AI the teacher can spot it from a 100 mile distance. Getting these kids to read anything longer than a sentence is almost impossible. So many of the other kids in his class have exceptionally low literacy rates and emotional problems (which seem to be brought about by the combination of Covid + iPad parenting) that I would argue AI usage is a low ranking issue. In fact I would go further and say if the AI can write something they are willing to read maybe we should just let them. The world for them won’t look anything like the world we grew up in and while (if true) I would be very angry that Google is pulling their support for students I think this isn’t that and that there is a strong anti ai political aspect to this story which I believe is overstated
I cannot stand the Chromebooks. I wish they would kill them off. My kids got them as well but they at least still use text books. They are not using them at home though. They got steam decks attached to a screen for their gaming computers and its made the install mods and other stuff actually learning some computing.
Can someone help me understand why can't the schools install another browser? Or default to another search engine? I thought that this was possible on chromebooks. While I understand that this may be a second best solution, can't the IT staff at the school boards implement these solutions?
My guess is that the IT staff at most schools is either not great or close to nonexistant.
Like teachers, they are overworked and underpaid, and, as you put it, close to nonexistant.
I used to work at a state college and during a major tech turnover we had 1 tech employee for 3 campuses across 3 counties. That 1 was me. 5000 students, hundreds of staff, and me keeping the lights on for 3 years.
The instant someone offered me $1 more an hour I quit.
K-12 IT employee here, I lurk HN a lot and made an account to throw in my 2 cents.
Anyway, both are generally true. My district uses a hub model, where support staff are dispatched out to schools and network infrastructure/business ops are centrally managed. But our model is common to larger districts; smaller ones often rely on small general-purpose IT teams - maybe a handful of people.
Obviously, being both public-sector and education, the pay isn't great because the funding isn't there. I like to think most of my colleagues are competent, service-minded people, we're just understaffed. We don't usually experience a lot of turnover, just growth from 2020-2024, but the economy has made some folks reluctantly move on lately. I'm hiring now and the strongest candidates withdrew when they realized what our salaries were. We're in a metropolitan area - rural districts have even weaker candidate pools and lower funding.
My district's leadership has embraced AI as part of our students' future, so they're taking an "allow it, but regulate it" approach and Google's change doesn't impact us. I have mixed feelings about how tech is integrated in grade schools in general, but this _is_ something of an added burden to workaround for under-resourced districts who don't want it.
Literally a one liner rule in ublock origin would block it district wide.
Unless I have missed something, the whole premise behind a Chromebook is that it's running...Chrome. And that's it.
Why not just use a different search engine.
Its not like there aren't other options. It seems unreasonable that a user who (presumably) isn't paying should be able to dictate product decisions. If you dont like google's decisions, vote with your feet.
We are living rimes, when workers have to advise people who lead people how much are social classes aparted from reality.
Google isn't doing anything that MSFT and Sun/Java didn't do before them.
What, so that makes it OK?
The only reason Microsoft didn't get broken up over this behavior is the travesty of the 2000 election coming just before the judge was going to apply the penalties, and the Bush Jr administration putting the kibosh on it.
That then, of course, set the stage (well, alongside the Reagan-era shift in enforcement) for the subsequent 2+ decades of kissing the boots of oligarchs and acting like massive corporate consolidation is normal rather than an absurd outlier that flies in the face of a century of antitrust law and practice.
I don't Google for answers as much as Google seems to think I should. I often Google to navigate the web but they are making this actively harder. Plus their LLM answers are frequently either missing the point or flat out wrong. I feel bad for those that will grow up taking the generative AI answers as verbatim because they do not know better. Though there is some hope, some kids I've talked to have been taught to distrust LLMs and to ignore or not trust AI results - so at least we have that.
I hope Google entrenches on the policy, and hope further that this causes total backlash to Big Tech in schools.
I haven't worked there in years but the visible heads of the AWU include some of the most clueless, dysfunctional people I have ever encountered and did not seem to me to represent anything resembling the central position of the typical Googler.
AWU is an unserious organization and absolutely does not have its act together. They seem ambivalent toward recruiting enough members to actually gain worker bargaining power (which like, isn't that the primary mechanism and purpose of a union?). Even before they have that power they want to charge you 1% of your salary to print t shirts and hold solidarity with contractors basically - the cart comes before the horse. This is literally just strongly worded letter without that power. Doesn't mean nothing will come of it (I hope it does much like the voluntary exit program), but if it is resolved, it will likely be due more to the larger outcry and only in very small part to AWU itself (again like the voluntary exit program). Very expensive socialist social club otherwise.
It might be better if schools used a search engine that was not developed by the world's largest advertising corporation.
And which would you propose? There's no search engine I know of that isn't run by some kind of profit-seeking corporation. And that's probably a good thing, since we all know what public sector funding for libraries ends up looking like.
Libraries are an amazing place, even with funding issues.
Wikipedia's seems reasonable. Certainly not ad-funded.
AI doesn't lobby their owners like this
In related news, Google punishes several employees for insubordination. Didn't they learn their lesson when this exact same thing happened with the Palestine cause?
“Punish.” As if schools moving away from Google products and towards almost anything else would be a bad thing.
It seems like a pretty simple idea for schools to just switch to ddg or bing if they choose to keep their students ignorant of AI until it replaces them.
Plus I don't get the point of petitioning a business to do something, it ain't a charity.
1. People petition businesses all the time, especially when the business has a monopoly. Besides regulations, things like the MPAA,ESRB, and even just Xbox backing down on some anti-consumer changes all are the result or (consequence if involuntary) of this.
2. Google absolutely behaving nominally charitably around schools. The strategy around things like Chromebooks may have a well troded ulterior motive but that isn't on the books.
1. What does google even have a monopoly on? Search? I just told you about ddg and Bing. About workplace suites you gotta be kidding me xD. Cuz there O365, Zoho, even proton has one.
2. When a business behaves charitably you just know they have some ulterior motives. I wouldn't trust them.
Unionize now
https://en.wikipedia.org/wiki/Alphabet_Workers_Union
They have one. The way that it was structured and the issues that it takes up make it a minority union and so doesn't have any bargaining power.
It's frustrating because you have to commit a non-negligible amount of your paycheck to the union and then it seems like this particular union (AWU) spends most of their effort helping non google employees (vendors, extended workforce, contract workers etc) negotiate what they want. I think if they thinned out their expenses to require lower fees and focused on larger issues it would be much more effective and have larger bargaining power.
As someone who spent a fair amount of time at Google trying to get people to join the union, this is 100% true. Almost everyone who I asked about it cited union dues as their main concern. Extremely frustrating.
How about forcing those enterprises pay for this? I think it's stupid but I'm not sure how
Union dues are paid by the members of the union.
Google's revenue doesn't go to a union (except by way of an employee electing to pay those dues).
Perhaps it would be easier for the union to get people to join if it did things that would actually benefit them instead of basically being a PAC for the omnicause, supporting things that directly oppose employee interests.
When I left Google like seven months ago, the union had decided (based on feedback from members) that it was primarily interested in job security related things, such as buyouts and severance packages. I don’t know if that’s changed or what.
Well, it certainly wasn't their concern for its first five years of existence, and its external communication only barely mentions job security for its members: https://www.instagram.com/alphabetworkers/
Dues are a whopping 1%. 1 whole percent. That's it.
For someone making hundreds of thousands of dollars a year, there’s a lot of sticker shock.
My heart. It bleeds for them. Truly, it bleeds.
They need 50% + 1 percent of the bargaining unit. Including others (vendors, extended workforce, contractors) in that not only makes it so that you need more people, but also that those individuals have conflicting goals which the union cannot represent (FTEs may have a workplace issue that conflicts with vendors).
As such, it cannot be a union registered with the NRLB and so will never have the bargaining power.
If it was instead a single group (e.g. subcontractors working for Google Fiber) then it would be easier to get the 50% vote and would have a single focus and could be a union that has collective bargaining power.
> have to commit a non-negligible amount of your paycheck to the union
Dues are 1%. How is that non-neglibile?
Well I guess 1% of pretax income is non-negligible depending on what you get out of it. If its a worthy cause sure, I probably donate closer to 2% every year to various things.
When it's a charitable donation to fund labor lawyers and organizers for employees of Accenture and Cognizant, its a no for me.
There are only so many slices of a pie
I mean, my favorite charities would be happy to accept 1% of your salary.
1% of salary seems like a significant amount. Either you're not making much and every dollar counts (but then you're probably not a direct Google employee), or you're making enough that 1% is real money that you could do something with.
SAG-AFTRA charges 1.575% on the first $1M of income plus a base of about $250, but it's an industry union with significant market power; membership comes with benefits to weigh against the dues and other drawbacks.
What benefits is AWU providing for members? How do they weigh against the dues and other drawbacks?
Any amount of money is non-negligible when it goes towards fighting against my interests.
lol, the AWU is a joke. As a sibling comment pointed out, it's a PAC disguised as a union. A quick look at the demographics of their members page and comparing it with the demographics of the average Google employee will tell you why.
Feel free to share which campaigns of theirs are political action: https://www.alphabetworkersunion.org/our-campaigns
I'd suggest looking at their original issues and the reporting around it.
https://www.theverge.com/2021/1/11/22224926/google-union-700...
https://www.wired.com/story/googles-new-union-addressing-pol...
----
It wasn't focused around workplace issues and instead was trying to force YouTube to deplatform Trump and have Google act in certain ways including the "control what we work on and how it is used" part.
AWU may have good and righteous issues... and some were workplace focused (termination and harassment)... but its largely been politically focused in its inception and messaging.
From HN back then:
A group of Google workers have announced plans to unionize - https://news.ycombinator.com/item?id=25630011 (1368 comments)
We’re Google Workers, and We Are Forming a Union https://news.ycombinator.com/item?id=25629852 (94 comments duped to the above one)
Maybe unpopular opinion, but unions should focus on improving working conditions, wages, and protecting workers from retaliation or unfair dealing not trying to influence the direction of the business. If a union were to push for the issue in TFA and it turns out to be the wrong decision, who is responsible? Will the workers take a voluntary pay cut to compensate for lost revenue? It also gives the executives another scapegoat if things do blow up.
>unions should focus on improving working conditions, wages, and protecting workers from retaliation or unfair dealing not trying to influence the direction of the business.
There is precedent for this in Germany, at least, though I don't know anything other than it exists.
https://en.wikipedia.org/wiki/Codetermination_in_Germany
In Japan (maybe other nations), unions can have a seat on the Board. They advocate for the employees, but as stakeholders in the enterprise, which I think is smart.
Not everybody knows this but you can actually have collective bargaining without the world ending.
I think the key difference for countries outside the US is they have stronger national cohesion and stronger national identity. The US is a country of immigrants (always has been, always will be, literally) and so it's harder to feel solidarity.
The US has an extremely strong national identity. The problem the US has is that everyone considers themselves to be a temporarily embarrassed millionaire.
The exploited underclass don’t see themselves as the exploited underclass, and so they have no desire to band together to improve their situation. They believe that they too will become rich if they continue to play the game they’ve been told to play.
> The US has an extremely strong national identity.
Polite disagree, and I say this as an American. The country is less than 300 years old. That's nothing.
I don't feel the US has an "extremely" strong national identity. It has a strong one, to be sure, but it is quite fractured.
In Germany as well.
"Maybe unpopular opinion, but unions should focus on improving working conditions, wages, and protecting workers from retaliation or unfair dealing not trying to influence the direction of the business."
That is the bulk of what union workers are doing every day through their unions.
That doesn't mean unions shouldn't try to intervene on issues of broader concern as well. It's not a zero-sum game.
> If a union were to push for the issue in TFA and it turns out to be the wrong decision, who is responsible? Will the workers take a voluntary pay cut to compensate for lost revenue?
If the executives make the wrong choice, it's already the workers who will make up for it with an involuntary pay cut or layoffs.
We’re no longer living in the 1930s or earlier. Unions can take on other purposes if the times require it.
It’s a bit like saying that unions should only be using typewriters because that’s what they always used.
Yes, but only if it’s ultimately related to working conditions.
Otherwise you haven’t joined a union, you’ve joined a PAC pretending to be a union.
Google _has_ a union, as others have pointed out. Unfortunately they're too busy virtue signalling and tilting at windmills i.e., doing everything but looking after employees' needs, which, AIUI, is the main purpose of an employee union.
Layoffs, forced "VEP"s, etc. are happening all over Google, but the union is more concerned about Gaza, Palestine, etc.
It’s almost like the union is just a distraction from the real worker problems.
i don’t understand, what is Google if not their employees?
i know i know, I’m being deliberately obtuse, but I just hate this notion that people are just employees and they throw their hands in the air or shift blame to management.
if you are an employee you are responsible to the eyes of the people
This seems like a reasonable stance and not an impossible one to defend. Google is making a bet that long term, everyone will use AI. I think that's undeniably true.
The friction here is that they don't want to bifurcate their main offers to maintain an "AI free" experience, because the thesis is this won't exist long term. I think this makes sense.
This is not opposed to telling employees not to "punish schools", whatever that means. Nobody is being punished, they're just not being catered to.
> everyone will use AI
A bet that everyone will only use AI. Which is wrong.
No, even if the schools enable AI, it doesn't always show up on search. But it's a bet on people using it far more than they do today, which seems right. We're already past the tipping point where frontier models are more helpful than search engines for information retrieval, and soon that will be distilled down into ultra cheap models and nobody except Luddites will keep insisting on doing searches manually.
I think the Luddites have a case because LLMs are irrelevant to the "search is your address bar" use case. Google took off when it became easier and faster than remembering the URL when you had a specific page in mind. The AI-summary contributes nothing to this except moving the link you want to click on several centimetres down the screen.
Those queries tend not to trigger the AI overviews, or they trigger below the first result and so do no harm if that's all the user wanted.
In my experience, those queries absolutely do.
I’ll type a one word search term to find a site, the result I want will be at the top, and Gemini will pump out a completely irrelevant half-page summary about something completely unrelated to what I was looking for.
Maybe don't frame everyone whose not a FOMO AI booster as a Luddite? Some of us want balance and nuance, and to treat AI as normal technology. It's mildly insulting and if you know your history, inaccurate.
Personally, I'm hoping that once the FOMO AI boosters throw around the term 'Luddite' enough it'll become a positive term praising people who carefully evaluate risks/benefits of new technology instead of thoughtlessly chasing the FOMO.
Luddites had a good political position, and their demands were reasonable. Of course, the capitalist classes used the power of the state to put them down with violence, because police exist to protect the property of the wealthy, not to actually support justice.
Do you feel that there is any inherent value in a school trying to educate a child and evaluate their learning outcomes?
No, Google has made a bet that forcing people to see AI search results will somehow be good for them. The moment the AI search results appeared, people asked for a way to turn it off. The answer was no.
They're not the only company refusing to let people turn off AI features people don't want. In this case, it doesn't appear to matter what the customers want.
I don't really see a problem here. Google has the right to design their product how they want to, and users have the right to stop using it if they don't like the product anymore.
The claim that "customers don't want this" isn't backed up by data. What percent of customers don't want it? What percent of customers actually do want these features and use them all the time?
That would have been nice before they obtained the entire market, but I guess school districts can just switch and retrain over the weekend? Schools have tons of money too so they can dump all those chrome books and buy iPads.
I'm pretty sure you can use Bing or DuckDuckGo on a Chromebook. There are also cheap netbooks available from other manufacturers.
Yes, and users of private railroads during the gilded age had the right to stop using them any time they wished.
I can further clarify, but I assume my point is clear.
That's a metaphor, not an argument. Search engines are not railroads. Google is not the only search engine.
My assumption of clarity was, apparently, unfounded.
Google has what would have been regulated away as a monopoly when there was an appetite for that kind of fight. It's an analogy, not a metaphor.
There are so few competitors because Google has systematically starved them or bought them, so that it is practically impossible for anyone to compete with them on product or reach. It is not because they are the only tech company capable of building a good search experience.
> The claim that "customers don't want this" isn't backed up by data.
You're pretending there is no data when we're here responding to an article about how some customers don't want it.
"Some" is vague and anecdotal. How many customers are we talking about?
Other people need exact numbers, but you don't for some reason?
Prove to us it's not a thing. Everyone would love to know this whole article is based on a false premise.
Here you go. More than 1 billion people are using the Gemini app every month. Presumably, those users find it useful, or they wouldn't be using it.
https://blog.google/innovation-and-ai/products/gemini-app/on...
Google Search has 5 billion monthly users. I would be surprised if anti-AI users were even close to 50M, i.e. it's likely far less than 1% of the userbase. But if you want to make that argument, I would be interested to hear it.
Using Gemini isn't the same as "happy that search puts AI summaries up top". "I would be surprised" addresses the point, but it's not "proof".
So you cede that there is data, you just want more specific data that can’t actually be collected
I cede that some percent of Google's users are anti-AI, because that's obvious. What I want is for the person making a claim to be specific about what they're claiming. Are we talking about 50% of users, 1%, or 0.001%?
You do realize that Google would be the only organization capable of getting that data definitively right? And they have zero incentive to ask the question?
A major municipal entity in a major municipality, we're talking about the entire population of their remit when we discuss customer impact. But focusing on something like 'percentage of total users' ignores the critical 'user value' dimension.
What they can’t do is sell a product, then fundamentally change that product instead of delivering the product was originally sold
Actually, they do have the right to do that. Companies do it all the time, that's how we get better products. Or they run the product into the ground and the users move on to better things. That's how the process of innovation through competition works.
> Google has the right to design their product how they want to,
Are we still arguing this in 2026? No they don't. Just shut them down.
Sounds authoritarian to me.
Google is an illegal spyware company. The authoritarian thing to do is to enable their operations.
I have to disagree on the punishing aspect. They enabled Gemini for student accounts and there was a significant backlash. They then added some configuration so that schools can disable AI features, stating that if all AI features are disabled then they will _also_ disable Google search.
If they didn't want it to sound punitive, another solution might have been more appropriate. Google search today has a "Web" item (listed under "More" at the top of the page) which does provide search reasults without the AI overview. I think this indicates that this feature isn't so difficult to provide to schools, their paying customers.
> the thesis is this won't exist long term
Well if that is the case, it would make schools’ decisions a lot easier if Google came out and said that.
Google's product is AI. It's like schools refusing to buy JIF because it contains peanuts.
Bread is flour, doesn't mean I want a cup of flour included with every loaf.
It's more like schools refusing to buy JIF because they switched from making peanut butter to mayonnaise.
Even that would be significantly more defensible for the manufacturer. They don't make it anymore, that sucks but gotta move on.
It's more like the bottom half of the jar is the same amount of peanut butter as ever and then they add a big squirt of mayo on top. They still make the old product as part of the bundle, and the bundle is the same price, but they insist on only having the bundle.
Which is perfectly sensible, both for the schools (get peanut butter from somewhere else) and the manufacturer (believes mayonnaise is more profitable than peanut butter). I don’t know why the manufacturer’s employees need to lobby the manufacturer to sell a product the manufacturer doesn’t want to sell.
Literally a CSS rule to hide the "AI overview" div can do what these schools are asking for
But google has always been an AI overview. Google filters out things you think don't matter - that's AI. Google judges what pages are useful and which ones are trolls - that's AI. Google analyzes your search and decides how to interpret the terms to bring you the most "relevant" pages in the context you're asking from - that's AI.
Google is AI to the very core, to the very marrow of its bones, and it always has been. It wasn't always llm, but to pretend Google isn't AI is just ignoring the statue and staring at the fig leaf and thinking that's the product.
It's a symptom of the fact that, for the specific use case of web search, the 'AI' Google has now is worse than the 'AI' Google had before.
We all keep stretching the term Artificial Intelligence, until it means nothing and everything, aren't we?
AI now encompasses we used to call “statistics”.
I hear parrots are called AI too now /s
this is pedantry. schools dont want kids talking to chatbots (I say this as someone that talks to chatbots all day). it's that simple
You know what "the AI overview div" means. Arguing about whether the rest of google counts as "AI overview" if you squint hard enough is a distraction at best. The div, the one at the top, the one with an LLM and a reply box, that's the div.
There are many things called "AI". Smart search ranking driven by machine learning that helps me find relevant original content written by real people is one. LLM-generated slop is another. More of the first, less of the second, please.
Google's product has become AI. It has to make it work because they sabotaged their other revenue streams with it.
What you're saying is that we'll eat it and we'll like it.
We don't have to.
Then I must have been mistaken when I clicked a search result and read my email earlier. I thought I was using Google products, but apparently not.