Google Search's 'udm=14' Trick Lets You Kill AI Search For Good (arstechnica.com) 40
An anonymous reader quotes a report from Ars Technica: If you're tired of Google's AI Overview extracting all value from the web while also telling people to eat glue or run with scissors, you can turn it off -- sort of. Google has been telling people its AI box at the top of search results is the future, and you can't turn it off, but that ignores how Google search works: A lot of options are powered by URL parameters. That means you can turn off AI search with this one simple trick! (Sorry.) Our method for killing AI search is defaulting to the new "web" search filter, which Google recently launched as a way to search the web without Google's alpha-quality AI junk. It's actually pretty nice, showing only the traditional 10 blue links, giving you a clean (well, other than the ads), uncluttered results page that looks like it's from 2011. Sadly, Google's UI doesn't have a way to make "web" search the default, and switching to it means digging through the "more" options drop-down after you do a search, so it's a few clicks deep.
Check out the URL after you do a search, and you'll see a mile-long URL full of esoteric tracking information and mode information. We'll put each search result URL parameter on a new line so the URL is somewhat readable [...]. Most of these only mean something to Google's internal tracking system, but that "&udm=14" line is the one that will put you in a web search. Tack it on to the end of a normal search, and you'll be booted into the clean 10 blue links interface. While Google might not let you set this as a default, if you have a way to automatically edit the Google search URL, you can create your own defaults. One way to edit the search URL is a proxy site like udm14.com, which is probably the biggest site out there popularizing this technique. A proxy site could, if it wanted to, read all your search result queries, though (your query is also in the URL), so whether you trust this site is up to you.
Check out the URL after you do a search, and you'll see a mile-long URL full of esoteric tracking information and mode information. We'll put each search result URL parameter on a new line so the URL is somewhat readable [...]. Most of these only mean something to Google's internal tracking system, but that "&udm=14" line is the one that will put you in a web search. Tack it on to the end of a normal search, and you'll be booted into the clean 10 blue links interface. While Google might not let you set this as a default, if you have a way to automatically edit the Google search URL, you can create your own defaults. One way to edit the search URL is a proxy site like udm14.com, which is probably the biggest site out there popularizing this technique. A proxy site could, if it wanted to, read all your search result queries, though (your query is also in the URL), so whether you trust this site is up to you.
For the 867th time today... (Score:5, Informative)
You're not getting answers "from an AI", really. It's RAG (Retrieval Augmented Generation) summarization. It's a very lightweight model trained and tasked only to summarize the top search results. It's not "the AI" telling you to eat glue, it's top search results telling you to, and the AI accurately reporting on them
If they want to prevent this, they'll have to let the AI act more independently rather than just summarizing whatever garbage comes up on a search.
Re:For the 867th time today... (Score:4, Informative)
If you don't like Google's UI, use startpage [startpage.com] instead.
It uses the results from major search providers including Google. It gives you a bit of privacy, too.
Or there is also DuckDuckGo [duckduckgo.com]
Re: (Score:3, Interesting)
It's a very lightweight model trained
Oooh trained model? I've heard of this. I think they refer to that commonly as AI.
Look I get what you're saying, but you're doing no one favours choosing this hill to die on. The world has evidently adopted the term "AI" to mean any algorithm which works based on a trained function. You have two choices, adopt the term as well and communicate to billions of people in an understandable way, or insist on being specific communicating with a couple of thousand experts who know what they are talking about, while
Re: (Score:1)
I didn't say it's not AI . I said you're not getting answers from the AI. The AI is only trained to summarize. You're getting a summary from an AI.
Re: (Score:2)
You're getting a summary from an AI.
Only if you're very lucky. LLMs are absolutely awful at summarizing text.
Sure, they're great at making text that looks like a summary. Just don't expect completeness or accuracy.
Re: (Score:2)
I literally train AI summarization models. They're *excellent* at summarizing text. It's one of the easiest tasks you can give them, and only takes a lightweight model.
Re: (Score:2)
I assume you do so as a hobby, not professionally. LLMs are notoriously bad at summarizing text.
Re: (Score:2)
I didn't say it's not AI . I said you're not getting answers from the AI. The AI is only trained to summarize. You're getting a summary from an AI.
And I didn't say anything that countered that either. I said "the term "AI" to mean any algorithm which works based on a trained function"
Look for what it's worth I agree with you. The problem is you and I talking on equal terms doesn't help the rest of the world that now (thanks to marketing bullshit) understand it to mean something completely different. The fact that you've put your foot down for the "867th time today" shows that you're on the wrong side of language as language is nothing more than a mech
Re: (Score:2)
The world has evidently adopted the term "AI" to mean [...]
Good for them. That's what the term has meant since 1956. (It's actually broader than that!) This is not a new thing.
This is a news story, not an PhD dissertation. AI is the "correct" term given the readership.
As it happens, 'AI' is the correct term in either case. What term do you think our imaginary PhD candidate should use instead?
I'm also more than a little curious about how you think the term "AI" should be restricted in a scholarly context.
Re: (Score:2)
LLMs aren't intelligent so like most "AI" it should probably called ML.
Re: (Score:2)
Machine Learning is a subset of Artificial Intelligence.
If it makes you feel better, according to Pamela McCorduck, there were people at the Dartmouth conference that weren't exactly happy about the term 'AI' either, but the time to object passed with the summer of 1956.
I understand that you would prefer to reserve the term AI for science fiction robots or whatever, but that use came long after.
Re: (Score:2)
That's what the term has meant since 1956.
Irrelevant. We're not speaking French. We don't have an equivalent of the Académie Française to manage our language and understanding. English evolves. Evidentially it has evolved in a way that you (and I) don't like. But we really only have two choices. Be a hopeless pedant pushing water uphill while angering those people just trying to have a conversation, or we can adopt the "new" meaning and use English for its actual purpose: communicate.
I'm also more than a little curious about how you think the term "AI" should be restricted in a scholarly context.
You completely missed my point. The point is not one of
Re: (Score:2)
Ah, I see. You've confused me for the idiot you first replied to who was making up his own stupid terms. I misinterpreted part of your post.
Re: (Score:2)
It's not "the AI" telling you to eat glue, it's top search results telling you to, and the AI accurately reporting on them
In any case, if you are eating glue because something on the internet tells you to, then I don't think the problem is on the other side of the connection ...
Or . . . (Score:3)
And I'm just spitballing here, you could not use that piece of garbage. I was forced to use it the other day because DDG wasn't returning anything. I went to Google and holy hell. I hadn't been there is years and didn't realize what a steaming pile it had become.
I can't imagine ever using that shit again unless I'm forced to. What used to be a decent search engine isn't even a shell of its former self.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
DDG is not a "wrapper over Bing", it is their largest but not only source.
https://duckduckgo.com/duckduc... [duckduckgo.com]
" DuckDuckGo leverages many sources, including specialized sources like Sportradar and crowd-sourced sites like Wikipedia. We also maintain our own crawler (DuckDuckBot) and many indexes to support our results."
https://www.searchenginejourna... [searchenginejournal.com]
" DuckDuckGo uses over 400 sources to provide results, including sources such as: Bing, Yahoo, Apple Maps, Wolfram Alpha, Yandex. They also use DuckDuckBot
Re: Or . . . (Score:4, Informative)
Re: (Score:2)
Re: (Score:2)
A dependency used in 0.00001% of cases will still prevent your code from compiling if the function were to disappear. You could say that makes it entirely dependent, you could also just replace it with a null function and have it mostly work. All we really know is that DDG is more than 0% dependent on Bing.
Re: (Score:2)
I tried DDG searches periodically throughout the Bing outage.
Initially they all returned a flat out error message, after a timeout. Literally no results. No way to "try again with ".
Later on there was a message about having problems, with hints about how to have DDG just go query Google (or some others I forget). For Google it was to start the query "!g ". I believe it then literally just performed a redirect to that other search engine.
So, yes, it does seem that in order to return any results DD
Re: (Score:2)
Yes, I see you agree with me. They are more than 0% dependent, and you have no way of knowing if it's 100% or 0.000001%
Startpage does not get all its results from Google (Score:2)
The increasingly leverage Bing. Indeed sometimes the results you get are identical to Bing. Don't believe they use Bing. Read their own page about this
https://support.startpage.com/... [startpage.com]
Re: Or . . . (Score:2)
DDG and Kagi and Brave are also bullshit due to AI.
Re: Or . . . (Score:1)
Go here to incorporate this into your browser (Score:5, Informative)
It's amazing and works well! It's like having Google from 2010. Go this web site ( https://tenbluelinks.org/ [tenbluelinks.org] ) and there's instructions on how to get your browser to use this as the default.
Re: (Score:3)
We have had that functionality (Google results without Google) for many years with StartPage.com:
https://support.startpage.com/... [startpage.com]
or
https://support.startpage.com/... [startpage.com]
Re: (Score:2)
I use StartPage too. This is different.
They are shifting to Bing (Score:2)
The mention it here
https://support.startpage.com/... [startpage.com]
Often their results are a carbon copy of Bing these days
Or... (Score:3)
>"Check out the URL after you do a search, and you'll see a mile-long URL full of esoteric tracking information and mode information [...]"
Blah blah, this site, that site, URL, jump through this hoop, edit that, try this, might last, can't make it stick...
Or you could just go to https://startpage.com/ [startpage.com] and get the same Google search results and with no junk. No AI, no tracking by Google, no animated banner annoyances, no asking to sign in pop-up, no "other people searched for" mess, etc. Why would you WILLINGLY want to give MORE information to Google, and be annoyed MORE on top of it?
Ah Startpage, that other BING front end (Score:3)
You do realise they increasingly shift to using Bing, right. They have been for a while and often their results are identical to DDG and Bing these days.
https://www.reddit.com/r/Start... [reddit.com]
https://support.startpage.com/... [startpage.com]
Re: (Score:2)
I have compared results on both Google and Startpage many times over the years. Just did it today. Searched for "testing". Looked through two full pages of results and they were exactly the same on both. This isn't proof they are ALWAYS the same, but it certainly looks convincing. Have you tried it yourself? Doesn't take long.
Certainly doesn't fit the "Bing front end" narrative. Besides, if they did have a blend (which I haven't seen), that wouldn't be a bad thing.
Another Search Engine? (Score:1)
Re: (Score:2)
*Literal* garbage, you say? ;)
nobeta=1 (Score:2)
I still use that when accessing slashdot. Wonder if it also works on youtube.
Kagi.com for paid search? (Score:2)
I've used it a bit and absolutely no BS, no ads, just direct hits.
enter the extensions... (Score:3)
For chrome, Firefox, edge, safari et al.
it is not AI that is the problem it is Google's UI (Score:1)
the google ui has become bloated with the cards etc it looks more and more like a web portal from the nineties, like yahoo. try the chatgpt app and you will get a clean ai interface.
Scaredy cats! (Score:2)
A proxy site could, if it wanted to, read all your search result queries, though (your query is also in the URL), so whether you trust this site is up to you.
What dosh! I'm sure that I can totally trust some "udm14.com" site with my data much more than I can trust one of the biggest, most scrutinized companies on the planet!