I just heard a Dutch person saying something about AI. I think this person has a point…
A quick translation:
“AI isn’t going to take your job. But the people who engage with it will.”
1 Like
Another good one, stolen from Twitter (Joanna Maciejewska):
“I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do my laundry and dishes.”
9 Likes
Or yet another one:
Q: AI will do everything for us. Working people will become obsolete in the future!
A: Wait, until you are old and need your diapers changed…
1 Like
one person has only 10 lines of informations to send. he uses AI to generate a 30 pages report to send to the coworkers… and the coworkers use AI to sum up the 30 pages to a 10 lines of usable datas…
6 Likes
That’s what happens often, Jean-Yves…
But I bet those 10 lines of usable data is not the same as the initial set of lines.
2 Likes
1 Like
My experiments with AI so far is:
It writes a lot of garbage, but in a so confident way that many could believe it could work correctly… If it could compile without errors or warnings.
I’m afraid of what will happen when it starts to compile insane wrong code with perfection and lazy people just start to copy and paste those things without understanding it.
1 Like
This is my experience also.
I tossed at AI two problems I have battled with for many years.
And I got so convincing results that I actually went ahead to try to try them, which took days to test (since those were complex problems).
The further i went the more I realized that it had tossed at me utter rubbish but put it in nice coat so it looked convincing.
6 Likes
It looks like AI is, at this point, the perfect companion to validate conspiracy theories. It can even create convincing new ones.
Maybe it’s good to write sci-fi stories. But the co-author must proof-read the output for plot holes.
There was a Brazilian character kind of called “Pantaleon” (Pantaleão) from a comedy show, that was a liar countryside man. He had stories of his adventures for any occasion, but if you interrupted him during his storytelling he changed things ending with some absurdity as a giant fish ending as a giant snake because he lost track of what he was talking about.
1 Like
Whilst we might mock it, we’re pretty much past the stage of being able to trust photos and videos that we see on social media. Some of them still have telltale signs, but I’ve seen ones that are almost impossible to tell if they’re real or not. When I see an image nowadays, I look to see when it was originally posted, so that I can rule out AI if its of a certain age!
Still, I don’t want to be a luddite about all of this. This isn’t the first huge technology shift. All this is going to happen whether we take advantage of it or not. But I do feel very uncomfortable about a lot of it.
3 Likes
Oh I do. I think you’ll find that the bad actors, whether they be state gangsters or private ones, just can’t wait to take advantage of this. Of course, they are already doing this with the whole social media scam, but this multiplies their ability to do so. Free elections, anyone? Forget it.
For sure, I don’t disagree. Yet, even if our own countries regulate against it in some manner, will that stop the bad actors? How do you stop it? I don’t think you can, so the rest of us just have to find positive ways to use it - and also teach kids not to trust anything they see on the net. It’s happening, whether we like it or not.
You know, the current stage of AI is that it really do not “think” as people may believe. It does a very complex collage of bits of information trying to fill in the gaps with made up probabilities based on what it saw before. That may end with interesting things or absurdities.
We will end with something great at some point in the future, I’m just not convinced that we are already there yet.
2 Likes
i am negatively charged 
where are the people who engage with nanotechnology or industry 4.0 hype …
we all make mistakes across the years which AI can’t solve.
it need a boss AI
which analysis,
project / code review,
nag for bad database design,
knowledge of business processes,
print a digital signed quitting
and do a better job (autonomous).
i use AI for birthday cards, looks somehow nice but it repeat and get boring,
did not process all the text input or misinterprets.
apple boast with AI and
i still wait that siri open youtube in safari and not answer that it not found the app youtube.
AI only works with a feedback and learning system.
AI is focused to be a license product.
It’s just evolving so fast. Images like you posted were everywhere last year. This year, with recent tools? Hard to spot that it’s AI, although I seem to have a kind of adverse reaction to some AI-generated images for some reason.
This is a good watch from Rick Beato.
1 Like
Music is easy, an unusual note inside a melody (because learning from past melodies how to build something similar is easy) will pass unnoticed, differently from a wrong number in a formula and the number of fingers of a person.
Musicians are really afraid of this.
My limited understanding of this is that it uses language models to train itself. What happens when bad actors have the same AI develop massive quantities of bad data and erroneous and false data to influence the language models to train them?
If AI is to evolve it must have good accurate data to train it. It’s impossible for someone to fact check the amount of data going into the models and therefore the generated AI is highly suspect.
I believe I read where one of the AI companies were signing a contract with Reddit to use their massive data set to help train their AI. If that’s the case, we’re doomed.
I’m waiting for someone to have placed a hard-coded password (or obfuscation code) into their code, then AI finds it and uses it to auto-complete in another app.
1 Like
Remember the MCP from Tron? I’m thinking Tron was a documentary. Just saying. 