FreeBSD AI Experiences

About that friendly and 'more welcoming then Linux' FreeBSD Community - while I most of the time agree with that sentence - there are cases about which I have been notified that sometimes admins/moderators of FreeBSD Forums are not very nice and RTFM is the best they can do.

While sometimes a good RTFM message needs to be delivered - please always tak a context into account.

It was really visible in latest AI/ChatGPT threads where people suggested ways to achieve more reasonable replies from AI bots - being messaged with RTFM only.

Its always good to read and understand FreeBSD man pages - but I also often use AI/ChatGPT for various cases - mostly to get a 'high level view' of something - and there is nothing wrong with using these AI tools or sharing how to make them work better in FreeBSD favor.

It would be great if all of us would remember that we all play in the same team here, some of us start the journey while some others are closer to the end of the path ...



Regards,
ver
 
I know some (most?) of You RTFM all the time - and using AI now seems for some like 'cheating' - but lets not confuse fast way to get 'some' answers - even not perfect - and the learning process.

Especially remembering how fucked up Google search engine is right now.

I personally needed to switched to DUCK.COM long time ago - because results from Google were just plain shit.

Its really HARD to get relevant information these days ... not only in FreeBSD land ... in ANY topic.
 
dear. vermaden :
what AI can we install it in freebsd ? and how to go to DUCK.COM ? In CN, we can't touch any more . that is my big problem. maybe other country was same . thanks. freebsd forums is good for me .
 
... and how to go to DUCK.COM ? In CN, we can't touch any more . that is my big problem. maybe other country was same . thanks. freebsd forums is good for me .
I wonder, how did you end up here? Via Baidu? Neaah, maybe Yandex? Must be Yandex, since Xi is so fond of that Atomic Midget.
We're in Off-Topic area, right? Right?
 
I wonder, how did you end up here? Via Baidu? Neaah, maybe Yandex? Must be Yandex, since Xi is so fond of that Atomic Midget.
We're in Off-Topic area, right? Right?
hi dclau:
i just want to study more about freebsd . that is very good OS for me . i am get forums from freebsd.org ,not baidu,yandex. ... i am nobody, don't know anymore XI. so can you show me what AI we can build it in freebsd14.1 ? thanks.
 
It's one thing to draw and process pictures, photos, sound, video streams for yourself via AI or to speed up the transfer and processing of arrays of numbers in non-critical areas. You can integrate AI into the search service for pets that have escaped from their owners. This is normal.
But where the issue concerns precisely critical LIFE SUPPORT systems... this is a serious test for AI :)

Also, for example, I do not need "INTEGRATION of a smartphone with AI into myself". Explanation: if technologies make my life easier, then tightly tape the smartphone to my water meter with scotch tape or electrical tape and let the smartphone pair with AI and draw graphs of water consumption, sort by day, by hour, convert reports to .csv and send reports to my email. If I need it, I will look at the statistics myself, without being imposed by the AI. But I absolutely do not need a smartphone with AI to drive me to nervous exhaustion with its "reminders", "improvements" 24/7, to be screwed to me through "dependence", "inevitability", "fashion".
Today teenagers sleep with smartphones. Teenagers constantly SORT, but DO NOT ANALYZE information.
This partly reminds me of sorting garbage. Systemic thinking is destroyed. Semantics is disabled. This is a tragedy.
 
I know some (most?) of You RTFM all the time - and using AI now seems for some like 'cheating' - but lets not confuse fast way to get 'some' answers - even not perfect - and the learning process.

Especially remembering how fucked up Google search engine is right now.

I personally needed to switched to DUCK.COM long time ago - because results from Google were just plain shit.

Its really HARD to get relevant information these days ... not only in FreeBSD land ... in ANY topic.
Did you ever try eTools.ch search engine? Or search.dismail.de?
 
dear. vermaden :
what AI can we install it in freebsd ? and how to go to DUCK.COM ? In CN, we can't touch any more . that is my big problem. maybe other country was same . thanks. freebsd forums is good for me .
What about various VPNs or Tor?
 
It's pretty useless and can barely be motivated. If someone asks for information that's clearly mentioned in any documentation, you can choose to ignore it or point at the doc anyway. Anything else is dishonest and very damaging to the progress of community-driven software development.
 
But where the issue concerns precisely critical LIFE SUPPORT systems... this is a serious test for AI
The real problem with (the current state-of-the-art) AI is that it can't EXPLAIN what it has done.

E.g., if a doctor recommends a particular treatment, he has some rational basis for his recommendation; one that his peers would likely support.

Imagine an AI "suggesting" a diagnosis or "recommending" a treatment to a doctor. If this makes the doctor aware of something that he might have overlooked, then great!

But, if the AI has "noticed" something (AI is "fancy pattern recognition") that it can't explain to the doctor, what then?

If the doctor ignores the recommendation and the patient has a "bad result", a lawyer could argue that the doctor had been advised (by the AI!) of this problem AND CHOSE TO IGNORE IT -- suggesting the doctor's fault in the patient's outcome.

If the doctor "plays it safe" and follows the recommendation -- but doesn't know WHY he is doing so -- how is that considered health CARE?

IMO, the best practical use of AI (and CGI) is to automate televised news broadcasts and weather forecasts -- and get rid of all those "talking heads" that add little value to the newscast. Imagine a CGI-generated "Max Headroom" character as your "nightly News Anchor"; his motions driven by an AI to mimic a "talking head" (how apropos!) while his speech is generated by another AI parsing the news feeds (that the human newscasters had previously parsed).
 
What about various VPNs or Tor?
Yes, that's the way out. But our Tor browser is broken now. The guy from China probably wants to tell us that his government has tightened the big iron nut through the Great Firewall of China. :) By the way, China is currently leading, in my opinion, in implementing social rating. If AI is integrated with calculating social rating, then that's wow...
Dear vermaden:
i can't get any vm to build a vpns to get out........for searching more questions. thanks.
Censorship is dictatorial. My government is doing something similar today: they are closing Telegram in universities and government agencies, closing social networks, etc.
The real problem with (the current state-of-the-art) AI is that it can't EXPLAIN what it has done.

E.g., if a doctor recommends a particular treatment, he has some rational basis for his recommendation; one that his peers would likely support.

Imagine an AI "suggesting" a diagnosis or "recommending" a treatment to a doctor. If this makes the doctor aware of something that he might have overlooked, then great!

But, if the AI has "noticed" something (AI is "fancy pattern recognition") that it can't explain to the doctor, what then?

If the doctor ignores the recommendation and the patient has a "bad result", a lawyer could argue that the doctor had been advised (by the AI!) of this problem AND CHOSE TO IGNORE IT -- suggesting the doctor's fault in the patient's outcome.

If the doctor "plays it safe" and follows the recommendation -- but doesn't know WHY he is doing so -- how is that considered health CARE?

IMO, the best practical use of AI (and CGI) is to automate televised news broadcasts and weather forecasts -- and get rid of all those "talking heads" that add little value to the newscast. Imagine a CGI-generated "Max Headroom" character as your "nightly News Anchor"; his motions driven by an AI to mimic a "talking head" (how apropos!) while his speech is generated by another AI parsing the news feeds (that the human newscasters had previously parsed).
Hmm... interesting approach. Wouldn't it be scary for us to live in such a "Healthcare" system? :)
I agree with your argumentation, but briefly about the "Health-burial" system (this is Ukrainian slang for the health system from the "Ministry of Health", which helps ordinary citizens to pass away to another world faster; to get rid of excess population). I will say right away that I am a moderate conspiracy theorist and THIS is NOT curable. :)
In Kievan Rus there was a concept of "healer", "medical work"... it is difficult to approximate this to English, because our languages have different vibrations. Nevertheless...
In the modern global world there is a concept of "medical practice", "medical doctor"... it is more about "business", about "systemic business". A modern doctor is already tied to big pharmacology, modern monitoring, smart machines, 3D analysis and modeling... If you add AI, the doctor will be in the COORDINATE SYSTEM of the machine. This is money, benefit, in some sense a CONVEYOR.
The era of humanity is leaving, which is being replaced by ALGORITHMS. Yes, modern surgery, filled with digital technology, 3D printers really saves patients' lives. A huge plus for this!
In my country, the so-called "Medical Information Systems" were introduced relatively recently.
There is a political lobby that IMPOSE by the force of a monopoly on the market of digital medical systems. This is helsi (TM). This is the level of Big Data. All this lives in the coolest DE NOVO data centers (Tier-3 level). Patients' personal data is now in commercial hands. Through commercial Oracle, patient databases from all over the country are replicated to databases in the United States and to beneficiaries. Unbelievable! A huge commercial non-governmental mobile operator in Ukraine bought this medical system. Simply gorgeous... the banking sector is also involved in this gang...
And in the protocol they wrote "your personal data" is not transferred to commercial structures.
And phones, names, place of residence, diagnoses, intimate details! EVERYTHING is transferred in order to
screw the patient into this system and pump out trillions of money. That's it.
With the help of AI, we will already be f*cked to the fullest. It will not be enough for them to simply create a mathematical model of your body. They need to know EVERYTHING about you, so that based on the mathematical model of the body and the disease database, the AI itself will prescribe medications and pills.
AI will know everything about your liver, kidneys, heart, brain.
AI will prescribe a miracle powder.
If something goes wrong... who to punish? HOW?

Yes, a LIVING doctor will soon become as unnecessary as a cashier at a supermarket.
Get the idea? :)
 
Back
Top