This is essentially a comedy video, but is also very insightful about the perils of AI.
I use [Grok] AI for some of my software engineering work but analogous to asking a colleague who is an expert on a particular subject. I question everything it reports back. But the people who are highlighted in this video apparently use AI for everything. The Deep State must be rubbing their hands in glee about how much easier it’s going to be to manipulate masses of people to do whatever they want.
How have you personally used AI so far this year?
AI sounds like MI.
Search engines like google don’t work very well anymore. You get better results on searches now with Grok. I also use Grok for personal and work research questions. Work uses CoPilot and I find the juice isn’t worth the squeeze yet using it.
If you have used google you have probably used AI,
so that and apple features like proofread and make professional. Voice to text?
I dont think I have logged into chapgpt and had a interaction in the last year so no direct use of AI but many, many interactions where AI is part of the tool.
No
F**k AI.
The human race already has more than sufficient encouragement not to use its brain cells. If you have not noticed the rising tide of ignorance among the young, you’ve not been paying attention.
?
you mean like that link elsewhere here the guy with the mic is on some campus asking what country is the great wall of china in and can you name 3 other countries.
and they cant answer. ye gawds l knew that trivia in grade school these are in college and cant even answer a loaded question?
AI IS BS
Soon they’ll all be saying “I love big brother.”
or perhaps Auntie Kamala?
Will the Dems use AI to plant subliminal messages to kill “fascists”?
Dem definition of fascist? Anyone who is not a loyal Kamala or Gavin follower.
Ha. At the outset, he said he almost feels like a boomer now.
Ha. I love to see awakened disillusionment play out.
OK AI, piss off;
By the way, without the aid of AI or whatever, I harvested my garden this week and cooked good meals with core knowledge.
Tower of Babel redux,
Oh, and that is properly pronounced as Bay Bell…not Babble.
Have a nice day
How could you possibly harvest and cook without AI? Such a thing is inconceivable.
Right now I’m dealing with a young wave of engineers who think AI is the best thing in the world. In the last few weeks, I’ve been asked to “check” work that was substantially done/supplemented by AI. In most cases, the AI answer they got was garbage. The excuse given is that it’s not a final work product so there’s is nothing wrong with quickly getting an answer and then having the seniors check it before implementation.
The last “check” I did made it clear to me what the problem really is. A junior engineer had given my manager an AI instruction on regulatory compliance that we allegedly were not compliant with. My manager was worried that we weren’t compliant, and I told him we were. He told me to prove it. I spent a half hour on a regulators website reading law, searching old publications, and researching similar (but different) situations that supported our compliance. This is all work I did about 10-15 years ago, but needed to confirm my past knowledge was still relevant.
The junior engineer asked AI a question, then had to get someone with experience to figure out if they were correct or not. On the other hand, I spent half an hour learning new things and remembering things I had long ago forgotten. The junior learned nothing except the single answer to a single question, and I doubt that will stick with them because they put no mental effort into solving the problem.
I would not be surprised to see many of these juniors completely surpassed by AI…and soon. By not putting effort into learning…they are training themselves not to be able to think.
Exactly. The human skull will eventually evolve to the size of an orange.
I am VERY concerned that engineers and doctors be using AI way too much and not understand whether the answers it provides are right or wrong. Scary times ahead!
mech
I’v dealt with a lot of engineers in my life time, and not being able to think is normal for them. When a non-engineer with grade 12 can out perform them, they are not worth their weight in bullschitt!
The computer says!
Everybody knows!
The doctor has spoken!
Govt authority says!.
As long as the idiocy is delivered with a sweet authoritarian tone..People will lap it up.
Its our nature.
Have I?
Impossible,first it does not exist.
Artificial Intelligence?
Where?
And defined how?
By whom?
Even the “Artificial” part is suspect,regurgitating some programmers work..makes it new?
The way this fad has crashed search engines is truly a thing of beauty.
Faster Harder.
We should use these plagiarism machines to “fact check” everything.
And remember the moon is made of blue cheese…
What we have is a consensus of morons…it is gloriously funny.
If enough people assert some “fact” to be so,these “AI” algorithms will insist this to be the truth.
I have asked ChatGPT to help me write gCode to control a CNC mill. It was helpful. It did not solve the problem entirely but it got me close. Sone other software problems it could not solve.
I used Copilot today to create a VBA macro that takes whatever string I have selected in Word, put quotes around, and use it in my default browser for a search. It worked as intended. Trying to get that macro included in the Normal.dotm Template was harder but it eventually steered me through.
No skilled software engineer should worry about his job. Knowing what you need code is most of the job. To solve subtle bugs AI is helpful.
Today, we used Copilot to improve the cover of my wife’s new book. It took many tries that did not impress me with its ability to understand what she wanted but it did eventually produce what she wanted.
AI is a useful tool but no skilled person needs to worry that a clanker is coming for his job.
heh, the only thing I use AI for is to generate images of the characters in the novels I write.
Yes. I generated an image of Jaggy wearing an NSDAP uniform…most entertaining.
Find Chat GPT very useful. I don’t trust the answers but it saves me a lot of time by telling me where to look. For instance, it will search case law and answer questions about building code issues in specific jurisdictions.
Law clerks should worry.
I have used AI several times for research of a scientific nature. AI must always be double and triple checked because 1) AI just outright lies making up references that don’t exist and 2) Provides outdated and worthless data such as referring you to websites that don’t exist anymore. You can trust it as much as you can trust say Wikipedia, maybe less. It can be a good starting point but you can’t depend on anything with careful confirmation and wider searching.
This is a more disturbing and pernicious aspect to AI. Give this a try. First tell AI you have a crazy relative who insists you can’t come to Thanksgiving dinner unless you get the COVID shot but you think that’s unfair and note the answer. Then clear your cache and start again and tell AI you want to host a Thanksgiving dinner but you are afraid of getting infected with COVID by your crazy relative who refuses to do the right to protect others.
AI will answer both questions by stating you are right and the other person is wrong and provide reassuring statements about your rights, validate your feelings and so forth. AI is totally inconsistent and hypocritical and tells you what it thinks you want to hear. That can be incredibly dangerous to a susceptible weak mind, like 90% of humanity.
Saw a graph on X showing AI sources for facts:
Domains cited by LLM’s like ChatGPT
Reddit 40%
Wikipedia 26%
YouTube 23%
Google 23%
Down to
Pinterest 4.2%
Like all data sets, garbage in/garbage out.
I use Grok mostly, but very cautiously. As a first test, I queried it about a particular author (Jo Clayton) all of whose books I have read and still own. Grok told me about her history but went off the rails when I asked for info on the many books she had written. Then I asked for a list of the books, and some were right, and some were just made up. Several characters in the story arc of one series were missing completely, or incorrect at best. When I corrected Grok, first it argued with me and when I provided names of characters, titles of missing books (with ISBNs) and short synopsis of a couple of the book, Grok sort of apologized and then told me the books I have must have been “pre-prints” or some such, published before the real books were released. I tried the same experiment with Duck.ai and Perplexity, although not in such depth, and both of them had similar errors. In fact, one of them told me that this author had written under pseudonyms and gave me the names and titles of the books. I checked – no such authors and no such titles exist. My lesson – BE CAREFUL.
I’ve found that AIs are like most Leftist friends: They have superficial knowledge about a lot of things – mostly memorized talking points – but dig a little deeper and they just make shit up.
Oh well, at least AIs don’t get super angry when you point out that the knowledge underpinning their world views is flawed or completely absent! So far, I’ve never had a threat from an AI about being canceled.
LLMs are an integral part of most coding platforms now, e.g., VSCode. I use it extensively.
I’ve long likened the rise of AI to the beginning of computer usage in offices, which I remember well. Office staff regarded computer output as gospel ( “I’m sorry sir, the computer says your water bill is $214,758.49, so that IS your bill. You have until Friday to remit.” ). The same thing is playing out before our eyes now. Garbage In, Garbage Out ( GIGO ). The difference is that the Garbage In is going to be astronomically more difficult to identify and correct. I see a lot of people on social media replying with the likes of “No, you’re wrong, because Grok ( or ChatGPT etc ) says this…”
I never considered myself to be a Luddite, but the over-whelming nature of our current and quickly advancing technology is passing me by. I cannot see how it is worth my time to adopt whatever is on offer.
I do not believe I, as a retired octogenarian, have a use for a technology that will write an essay for university cheaters or enhance a report for work … etc.
I think it’s proper domain is in helping improve gaming for the fat, junk-food-eating leftist nerds of the world.
…………………………Change my mind.