I use spellcheck but hate-hate-HATE #autocomplete & Siri.

They only remind me tech is 90% based on assuming we’re stupid and further dumbing us down, the word I want so often not there if not actively avoided.

Still, there are some annoying #typing tasks where autocomplete saves effort.

So here in two posts are my 'Text Replacement' entries in IOS Settings.

An example is: I type 'isnt' and Text Replace offers 'isn't'.

1/2

Autocomplete-Fail des Tages:

ErfrischungErfrierung

"Nass aber heil zu Hause angekommen Bei den sommerlichen Temperaturen eher eine angenehme Erfrierung! "

Your #AI "is an incredibly sophisticated form of pattern matching, so advanced that it can mimic the output of human reasoning for a narrow band of problems. But when tested in a controlled way, its fragility is exposed. It lacks the robust, generalizable, and symbolic logic that underpins true intelligence.

We’re seeing the limits of very expensive #autocomplete that breaks when it matters most."

https://ninza7.medium.com/apple-just-pulled-the-plug-on-the-ai-hype-heres-what-their-shocking-study-found-24ad42c234a0
https://archive.ph/ASo9a
https://ml-site.cdn-apple.com/papers/the-illusion-of-thinking.pdf

AI can't do your job
"By firing #skilledhumanworkers and replacing them with spicy #autocomplete, #Musk is assuming his final form as both the kind of boss who can be conned into replacing you with a defective chatbot and as the fast-talking sales rep who cons your boss. Musk is transforming key #gov functions into high-speed error-generating machines whose human minders are only the payroll to take the fall for the coming tsunami of #robotfuckups."
@pluralistic
https://pluralistic.net/2025/03/18/asbestos-in-the-walls/

Okay, one more time for the people in the back.

The "AI" () craze of the past few years is all about Large Language Models. This immediately tells us that the only thing these systems "know" is trends/patterns in the ways that people write, to the extent that those patterns are expressed in the text that was used to train the model. Even the common term, "hallucination," gives these things far too much credit: a hallucination is a departure from reality, but an LLM has no concept of reality to depart from!

An LLM does exactly one thing: you give it a chunk of text, and it predicts which word will come next after the end of the chunk. That's it. An LLM-powered chatbot will then stick that word onto the end of the chunk and feed the resulting, slightly longer chunk back into the model to predict the next word, and then do it again for the next, etc. Such a chatbot's output is unreliable by design, because there are many linguistically valid continuations to any chunk of text, and the model usually reflects that by having an output that means, "There is a 63% chance that the next word is X, a 14% chance that it's Y, etc." The text produced by these chatbots is often not even correlated with factual correctness, because the models are trained on works of fiction and non-fiction alike.

For example, when you ask a chatbot what 2 + 2 is, it will usually say it's 4, but not because the model knows anything about math. It's because when people write about asking that question, the text that they write next is usually a statement that the answer is 4. But if the model's training data includes Orwell's Nineteen Eighty-Four (or certain texts that discuss the book or its ideas), then the chatbot will very rarely say that the answer is 5 instead, because convincing people that that is the answer is a plot point in the book.

If you're still having trouble, you can think of it this way: when you ask one of these chatbots a question, it does not give you the answer; it gives you an example of what—linguistically speaking—an answer might look like. Or, to put it even more succinctly: these things are not the Star Trek ship's computer; they are very impressive autocomplete.

So LLMs are fundamentally a poor fit for any task that is some form of, "producing factually correct information." But if you really wanted to try to force it and damn the torpedos, then I'd say you basically have two options. I'll tell you what they are in a reply.

@nytpu @claudiom @kelbot @gemlog @goosey @ParadeGrotesque @dibi58 @Maya Just wanted to share my #AI #autocomplete demo on my 20 years strong #Palm #Z22 #PDA! It works great. https://youtu.be/qJlHQcT-3S4

While google does propagate #materialdesign the Chrome browser shows an #autocomplete d #password field always in #urineyellow and you can't override cause the @paul team decided that this is "!important" -

however, in #CSS :autofill you can use hue rotate and other filters.

E. g.
filter: hue-rotate(230deg);
adjusts it to my violet color schema.

Biddde.

I'm curious.

Log into YouTube search for "How To Use A" and then tell me what the first ten auto- complete items are in a reply to this toot - I'll go first:

I've said that I estimate copilot to be the only profitable genAI service, but that might change ...

#Apple advertising their own apps in the built-in dictionary!?

As a #dyslexic #author, the advent of #spellcheck and usually #autocorrect / #autocomplete has allowed my work to read much more professionally. I rely on the phone reading to me because diary and dairy look like the same word, though I know better. Today, I decided to use the Look Up feature to check if I got the right word.

This screenshot is what I found. Incidentally, I spelled the word right as you can see from context.

#BoostingIsSharing

Little web dev tip: if you have just one address field in your HTML form (e.g., a textarea) but you still want the person filling it in to avail themselves of autocomplete, here’s a little reusable snippet you can use to achieve that:

https://codeberg.org/aral/gists/src/branch/main/single-html-address-field-with-autocomplete-using-javascript.md

(Ideally the web spec should be extended to include a full-address value for the autocomplete attribute.)

#DoctorWho #Autocomplete

In Doctor Who tonight, the Doctor and Donna will meet in a bar at a restaurant and I have a question for the audience and the Doctor and Doctor and Doctor and Doctor are both very interested and very excited to meet in the future together in a very exciting and interesting show.

#AI #Google #Spellcheckers #Autocomplete #Writing: "Soon after ChatGPT was released, an artificial intelligence researcher from one of the big tech companies told me that I shouldn’t worry about how the technology would affect how students learn to write. In two years, she assured me, only aspiring professional writers would enroll in writing classes; no one else would need to write anymore.

I remembered that conversation recently when I got advance access to Google Docs’ new “Help me write” feature, which is expected to be rolled out to all users soon. Once you have this feature, it becomes the default in Google’s word processor: The magic wand appears every time you open a document, ready to generate and revise text for you. If you want to write yourself, you have to close the feature."

Oh thanks #Google #Spreadsheet #autocomplete

well my plan was to use a simple * ...but if you think that's better...
#autocomplete #zsh https://social.tchncs.de/media/6BBSaBlXFR9cWYLktFQ