The full list: https://swipeblocked.com/list
Created by the one who discovered the list, Derik Kauffman
what the fuck
I know I’m naive about some things, but this is just… the fuck. changing the word before the most recently swiped word, because together it forms a blocked phrase? that’s just fucked up.
I’m ashamed to say that I truly never thought about the keyboard no longer being safe from interference.
like okay I get that it won’t suggest harmful phrases, sure, maybe I can accept that. but to retroactively edit your inputs… that’s just fucked up.
I do feel like autocorrect has been getting consistently worse over the last 2 years.
+1 to this. Tried a bunch of FOSS keyboards and switched to HeliBoard because of it! Wouldn’t you know, autocorrect experience improved 😮💨
Microsoft used to have a word flow keyboard I absolutely adored and they killed it. I miss it terribly.
Sounds like BS. 100,000 words and phrases? Looks like more commonly written phrases always win over similar, less common ones. You can probably find millions of such less common phrases. It doesn’t mean they are actually listed somewhere in the system.
Did you read the article? It says he found the list
Disgusting. I always turn it off when I can find the setting, it should not be the default.
I wish all phone keyboards had a easy toggle to turn off all “smart autocorrect” features so it types exactly what you entered with no changes when you want to
That’s already how it normally works if you just press buttons. This is for swipe-typing, which has to be “smart” because you swipe across many buttons and it has to figure out what you meant.
I fucking hate the policing of words by admin dickheads and corporate censorship algorithms online. Policing other people’s language flies in the face of every freedom and right people have fought and died for. Also the internet is the easiest place to ignore people, give people who care the tools to filter their feeds and comments instead of blanket banning everyone, cause it might upset some vocal minority
Who the fuck do these people think they are, deciding what we can and cannot say, what we can and cannot express?
I never really understood the whole “No really, we HAVE to guillotine them all” attitude of the 18th century French until I met tech bros.
I think you’re being a little reactionary, it’s more about offending someone by suggesting these things when they WEREN’T what they were actually trying to say. Autocorrecting to insults or swears could get someone in potentially serious trouble. It’s easy to imagine the viral facebook post “My iPhone got my fired!” Or even “iPhone is teaching my kids bad words!”
You cant have control of your own device, it could be a slight PR inconvenience for the company!
On 4th of June, 1989, in Tiananmen square, Beijing, nothing particularly interesting happened.
What most people call “algorithms” are not algorithms, artificial intelligences have replaced math formulas to back the recommendations your phone makes to you. Welcome to techno-feudalism!
that’s literally algorithms lol.
Algorithms are discrete operations done in a sequence, they are typically blind to ideas or personal opinions and are used for data processing, filtering, or calculations. Facial recognition algorithms are typically what you would call a complex algorithm as it uses conditions based on what the camera captures to make decisions as to where the face of said person is, if it is a person, are they smiling? What are their physical traits? Do they have white, brown, or dark skin, to adjust the contrast level for a good photo.
Artificial intelligence models are not math driven. They are language models, trained through machine learning and operating on a transformer architecture. Not only do they not replicate the reasoning a human (they predict the probability of action based on statistics of choices of others) and are most definitely going to be used to influence choices and opinions, not in an educational manner but in coercive and controlling manner. And it will become very hard to understand what AI produced propaganda will become. Even scarier though to me, is that it will be even harder to understand the effects of AI use, since all political actions have a direct effects which are often easy to predict, but they also cause these “ripple effects” which are the more dangerous side of politics and these ripple effects caused by AI use in politics will be impossible to predict for me, you and the AIs due to their early stage.
Algorithms are based on logic whereas LLMs are based on learning. LLMs have a layer of semantic understanding, which makes them more powerful.
laughs in sw engineer.
Meanwhile I, systems engineer wonders why the IT dude called himself an “engineer”
it dude? lol. i make robotics. including computer vision and ml. worry not. i’ll think of you every time an algorithm gets too many layers and starts to smell like an ai.
Computer vision and machine learning has little to do with robotics, but alright since you seem to work with know something about thermodynamics and are not a sys_admin like all the desk jockeys, i ll be nice and call you the E word.
But robotics itself is such a small niche that random people invoke because they see it in movies that I am skeptical to believe it as a systems and automation engineer, that s basically as close as 95% of engineer will ever get to robotics unless you re in China (mechatronics is not robotics btw) or work for Kuka/ABB…
lol. say that to my 6 axis machine that picks and pours and places.
bad bot. try again.
Artificial intelligence itself is a series of math formulas on very big vectors though.
Apple’s autocorrect is the best example of enshittification that exists to date. It was exponentially better 10-15 years ago.
It’s probably good if my phone doesn’t call my mother a cunt while telling her I’m bringing the cocaine by default, but I should be able to toggle that.
just give it to me, and i’ll bring that cunt her blow
Of all the headlines they could have picked…
In case you didn’t read, this is for swiping. If you type the example listed “United Ireland” that is okay. Try swiping that same thing and you are autocorrected to United Iceland.
I don’t know anyone that swipes.
Hi, me, I do a lot of swipe input. On Android, though.
I guess I’m weird but I can’t imaging not swiping. How did it not win out as the default input method?
Not weird. I think it must be a specific age group that tap-types faster than they swipe.
Yeah, I don’t have patience to type if it’s not a real keyboard. Swiping is what I’ve done for years on my Android phones.
Apple autocorrect and suggestions are pure shit. Especially on the touch swipe keyboard; it will just straight up make up words
I think it got even worse when they went to “AI” (Aborted Intelligence) in 2024, and they just have too much hubris to fix it
It shouldn’t be like that, but go into the dictionary and autocorrect whatever it tries to change it from, or make an abbreviation etc. I forced mine to stop saying “duck” instead of the fuck word.
the fuck word.
Oh? And which word is that, exactly? hehehe
It’s kind of funny. Some years back when I used to use Gboard I had to opposite problem
I wanted to text a friend to meet at the Golden Nugget Casino.
It has a bigger parking lot and a buffet.
Gboard would try to change those words to racial slurs. Had to go into the dictionary override settings and specifically delete certain words so it would never suggest them
You know Gboard uses all your typing history to make suggestions, right?
(just teasing you)
