Open Thread - 02-17-23 - AI and GIGO
The term artificial intelligence (AI) is being bandied about with regularity nowadays, especially in regards to chatbots, with ChatGPT being the most prolific. Hopefully this piece will help explain the concept of AI in context of its current applications and what that may portend for the future. Included, at the end, I'll let my mind run wild in a thought experiment regarding AI's potential abuse.
Artificial Intelligence
What is AI and how does it apply to chatbots and what are the different types of AI.
Artificial intelligence is the simulation of human intelligence processes by machines, especially computer systems. Specific applications of AI include expert systems, natural language processing, speech recognition and machine vision.
...
In general, AI systems work by ingesting large amounts of labeled training data, analyzing the data for correlations and patterns, and using these patterns to make predictions about future states. In this way, a chatbot that is fed examples of text chats can learn to produce lifelike exchanges with people, or an image recognition tool can learn to identify and describe objects in images by reviewing millions of examples.
...
What are the 4 types of artificial intelligence?
- Type 1: Reactive machines. These AI systems have no memory and are task specific. An example is Deep Blue, the IBM chess program that beat Garry Kasparov in the 1990s. Deep Blue can identify pieces on the chessboard and make predictions, but because it has no memory, it cannot use past experiences to inform future ones.
- Type 2: Limited memory. These AI systems have memory, so they can use past experiences to inform future decisions. Some of the decision-making functions in self-driving cars are designed this way.
- Type 3: Theory of mind. Theory of mind is a psychology term. When applied to AI, it means that the system would have the social intelligence to understand emotions. This type of AI will be able to infer human intentions and predict behavior, a necessary skill for AI systems to become integral members of human teams.
- Type 4: Self-awareness. In this category, AI systems have a sense of self, which gives them consciousness. Machines with self-awareness understand their own current state. This type of AI does not yet exist.
(Bold text added by the author)
Garbage In and Garbage Out
Garbage In, Garbage Out (GIGO) is a term that illustrates the process of computer data input and how it relates to data output. The data output quality is only as good as the data input. In other words, computers rely on the information that it's given and its output depends solely on the input.
GIGO (garbage in, garbage out) is a concept common to computer science and mathematics: the quality of output is determined by the quality of the input. So, for example, if a mathematical equation is improperly stated, the answer is unlikely to be correct. Similarly, if incorrect data is input to a program, the output is unlikely to be informative.
George Fuechsel, an early IBM programmer and instructor, is generally given credit for coining the term. Fuechsel is said to have used "garbage in, garbage out" as a concise way of reminding his students that a computer just processes what it is given. The term is now widely used in computer science classes, IT services and elsewhere. In fact, GIGO is sometimes used to refer to situations in the analog world, such as a faulty decision made as a result of incomplete information.
A variation on the term, "garbage in, gospel out," refers to a tendency to put unwarranted faith in the accuracy of computer-generated data.
(Bold text added by the author)
True AI and the Singularity
Computers do not think on their own, yet, and may never cross the sentient level of consciousness called the singularity.
What is the singularity in technology?
Most notably, the singularity would involve computer programs becoming so advanced that artificial intelligence (AI) transcends human intelligence, potentially erasing the boundary between humanity and computers. Nanotechnology is perceived as one of the key technologies that will make singularity a reality.
This intelligence explosion will significantly impact human civilization. These computer programs and AI will turn into superintelligent machines with cognitive capacity beyond human capabilities.
(Bold text added by the author)
Machine Learning
Machine learning is a more accurate descriptor of how current AI learns, it cross indexes a programmable database.
What is machine learning?
Machine learning (ML) is a type of artificial intelligence (AI) that allows software applications to become more accurate at predicting outcomes without being explicitly programmed to do so. Machine learning algorithms use historical data as input to predict new output values.
...
There is also the problem of machine learning bias. Algorithms trained on data sets that exclude certain populations or contain errors can lead to inaccurate models of the world that, at best, fail and, at worst, are discriminatory. When an enterprise bases core business processes on biased models it can run into regulatory and reputational harm.
(Bold text added by the author)
LMMRW (Letting My Mind Run Wild)
Let's imagine that there's an entity, a government for example, that wishes to control its subjects (heh, imagine that), and this entity may use AI to meet that objective. It need not matter if it's true AI or machine learning AI if the entity can convince its subjects that it's sentient and has access to all the knowledge of the known universe. A strong marketing campaign could be released upon the subjects to convince them that AI is the be all/end all of human knowledge and anything the AI says is therefore gospel. Keep in mind that AI is only as accurate as the data that is fed into it, in its current iteration that is, and is only capable of thinking on its own in limited terms.
Using biased programming an entity could resolve any issue by deferring to AI in what may be regarded as the sum total of all that is known within the scope of human knowledge. And then as a result the AI decisions must be adhered to because, well, it's AI, and it's smarter than us. Or so the story line may go.
An entity could theoretically control the world using AI, sentient or not.
Need an example?
And I agree, artificial intelligence, but not only artificial intelligence, but also the metaverse, neospace technologies, and I could go on and on…synthetic biology. Our life in ten years from now will be completely different, very much affected. And, who masters those technologies, in some way, will be the master of the world.
Klaus Schwab - February 14, 2023 at the World Government Summit
True sentient AI is a long way off, IMHO, although there may be entities that will try to convince you otherwise in an effort to meet their own objectives. The current AI is not sentient, it's GIGO. If the data input is biased the data output will reflect that bias.
If an algorithm comes along, you must whip it. Now whip it good. Into shape. Shape it up. Get straight.
Comments
Good morning Free Rangers...
things were so much simpler back in the primeval days of analog.
Tesla recall
https://www.cnn.com/2023/02/16/business/tesla-fsd-recall/index.html
語必忠信 行必正直
Tesla...
debugging their software using humans as Guinea pigs and a US government agency doing their job. The former is not surprising while the latter is and may be fallout from the twitter revelations.
Thanks soryang.
AI is making its' way
into the legal profession. For example, attorneys pay for a legal search of data to determine the likelihood of an appellate court outcome before they will take their case up on appeal. Who, what, how would you know if that data is GIGO ?
There are also "exciting" plans for making AI programs for representation of indigent criminal defendants. So, the days are coming when instead of appealing a prison sentence based on incompetency of one's attorney will be a fight between the State's data system against the people's data system.
Will law school become a data programming education one day?
I hope to retire before that shit starts flying through the courtroom.
Danger! Danger, Will Robinson!
"We'll know our disinformation program is complete when everything the American public believes is false." ---- William Casey, CIA Director, 1981
I can just imagine the day...
when an AI robot attorney malfunctions and an innocent victim is found guilty. Will that be considered a software glitch? Or maybe someone triggered the corruption algorithm?
Will Supreme Court
"We'll know our disinformation program is complete when everything the American public believes is false." ---- William Casey, CIA Director, 1981
AI is intruding into lots of industries and it’s being created
to make 90% of us useless eaters. Musk recently told us that when he spoke about his plans for the future. Right now he is working on creating a chip to be installed in our brain so that we can be hooked up to computers through WiFi. He is torturing thousands of monkeys now and most of them have died. Animal testing should be outlawed. A civilized society would have done that long ago.
The ideal subject of totalitarian rule is not the convinced Nazi or the dedicated communist, but people for whom the distinction between fact and fiction, true and false, no longer exists.
~Hannah Arendt
Timely essay
Is a Robot About to Take Your Job?
The ideal subject of totalitarian rule is not the convinced Nazi or the dedicated communist, but people for whom the distinction between fact and fiction, true and false, no longer exists.
~Hannah Arendt
My doctor shows graphs and charts
"We'll know our disinformation program is complete when everything the American public believes is false." ---- William Casey, CIA Director, 1981
AI
will come into it's own when the wealthy can buy a machine to play the market and the world economies to get wealthier, and then fire all those Harvard MBA's. Then there is only the politicians to have to pay off, for now.
Those machines...
already exist, but I catch your drift.
Yeah man, firing all those Harvard MBAs, I can live with that.
AI will probably have some kind of ransomware coded into it that will demand the politicians be paid off, or no results. That will be the corruption algorithm I mentioned above to otc. The vig will be written into it.
in other words it is artificial and therefore simply idiotic
Intelligence in, artificial idiocies out, is not what I consider intelligent. I like my gut's intelligence, if you can stand the smell, but so far it worked well.
Thanks for still being here. Hope all is well around your woods.
peace and love to all.
https://www.euronews.com/live
Accurate...
analysis, mimi.
Since I live in the Texas Pineywoods I can safely say that all is well in my neck of the woods.
You have some interesting wildlife animals in your
Pimey Woods area-
What is a Bobcat and what is a cotton mouse, sounds cute. No need to answer, it is too late in my part of the woods and I am going to sleep.
Good Night.
https://www.euronews.com/live
Bobcats and cotton mice
"Bobcat" is a short-tailed wild cat native to North America, a member of the lynx family.
"Cotton mouse" is a largish mouse that likes to use raw cotton to build its nests. Do not confuse with cottonmouth snakes, which are not cute at all!!! (They eat cotton mice.)
There is no justice. There can be no peace.
If they had self-awareness - I think they would kill themselves
immediately,
"Self-awareness. In this category, AI systems have a sense of self, which gives them consciousness. Machines with self-awareness understand their own current state. This type of AI does not yet exist.
Would they also have morality?
Oh boy, it gets sorta complicated to think about that. Good that I stopped thinking long ago.
https://www.euronews.com/live
If it comes to that point...
I would feel much better that they kill themselves rather than us.
I would venture that AI morality will be on par with human morality, some will be givers and some will be takers, that's another reason that I hope that the singularity doesn't become reality.
This article might be the bedrock for what AI
is being used for
I found this to be the money paragraph....ymmv
https://www.uclftr.com/post/blackrock-aladdin-intelligent-finance-why-ai...
bolded was my emphasis, as stated earlier GIGO, history doesn't
always repeat, especially regarding financial commodities
I never knew that the term "Never Again" only pertained to
those born Jewish
"Antisemite used to be someone who didn't like Jews
now it's someone who Jews don't like"
Heard from Margaret Kimberley
Oh stop it, g...
you're going to make Larry Fink salivate.
Now we know who they're talking about when they say that AI will benefit mankind.
Well if this is how, I'll take it ;:-)
I never knew that the term "Never Again" only pertained to
those born Jewish
"Antisemite used to be someone who didn't like Jews
now it's someone who Jews don't like"
Heard from Margaret Kimberley
As a side note:
More than 50 percent of the yearly US GDP is now derived from "financials" — rather than production or services.
Making money out of manipulating money or exploiting asset valuations.
This is the preferred occupation of our overlords, who have successfully objected to paying income tax on this type of earnings. They prefer to see income taxes derived exclusively from labor and wages.
Hola Johnny. Arguably it will always be GIGO because
they are trained on human product/output.
Wanna create a WTFbot? Train a neural net by feeding it the Congressional Record and a selection of national and regional Daily and Weekly newspapers starting with the debates and discussions leading to the creation of the Articles of Confederation. Jeebus whatta mess.
Then there's the problem with outcomes. Who decides whether an outcome is good or bad and at what time and place and for what duration? How to teach that fallacies are invalid but can, all the same, generate a correct conclusion:
"You can tell if an electrical outlet is functioning properly by sticking a hairpin in it" is a true statement.
be well and have a good one
That, in its essence, is fascism--ownership of government by an individual, by a group, or by any other controlling private power. -- Franklin D. Roosevelt --
Hola mi hermano...
you just described the current subset of political cretins inhabiting DC:
Ahah, that's how we can thwart the robots:
Have a good one. old buddy.
you too compadre
That, in its essence, is fascism--ownership of government by an individual, by a group, or by any other controlling private power. -- Franklin D. Roosevelt --
the human mind vs. artificial intelligence
common sense -
the basic level of practical knowledge and judgment that we all need to help us live in a reasonable and safe way
critical thinking -
the process of thinking carefully about a subject or idea, without allowing feelings or opinions to affect you
who the hell is going to create those algorithms?
think for yourself
[video:https://youtu.be/vtx5NTxebJk]
question everything
Heh
The bot must figure them out itself. When you internalize an externally created algorithm as a fixed and inviolable rule you have a religion.
be well and have a good one
That, in its essence, is fascism--ownership of government by an individual, by a group, or by any other controlling private power. -- Franklin D. Roosevelt --
Right, just what we don't need is another religion
logic -
a particular way of thinking, especially one that is reasonable and based on good judgment
analysis -
a detailed examination of anything complex in order to understand its nature or to determine its essential features : a thorough study
question everything
Those attributes...
wont be needed after humanity is enslaved.
Besides, there isn't a lot of common sense and critical thinking among humans nowadays either.
Thanks Q.
yo, that's all been subbed out to
info bots, so we can have more mental leisure time
question everything
Interesting essay on biases
Especially in these days of wokeism.
How smart people believe stupid things
And don’t forget the people who put Ukraine flags in their bios.
The new thing that drives me nuts is saying that men can get pregnant. Now instead of saying pregnant women, the thing to say is pregnant people. Schools are putting tampon dispensers in men’s restrooms. Gah!
My opinion is that this new wokeism thing is just a farce for further dividing the people so we fight over whether men can actually get pregnant instead of seeing how the parasites are robbing us blind.
I’m constantly asking myself what do I believe is the truth, but is just my conformation bias? I know that I believed a few things 3 years ago that I no longer do, but….?
It’s funnier than hell that government has decided that it’s the arbiter of what misinformation is when it has a long history of actually lying to us. Don’t you think?
More
The ideal subject of totalitarian rule is not the convinced Nazi or the dedicated communist, but people for whom the distinction between fact and fiction, true and false, no longer exists.
~Hannah Arendt
All of those...
are devices to keep us dazed and confused and divided.
With that line of thought, I think one of the biggest divisions in this country is between those that trust the government and those that don't. They don't talk about that one too much.
Thanks snoops.
The essay calls out the people who used to not believe
the government and the corporate media, but now they believe every word told them. They knew that the reasons for the Iraq war were bogus, but now they believe everything they are told about the Ukraine Russia conflict and its because they believed everything told them about Russia Russia…
The ideal subject of totalitarian rule is not the convinced Nazi or the dedicated communist, but people for whom the distinction between fact and fiction, true and false, no longer exists.
~Hannah Arendt
AI my foot
Artificial intelligence is not to be confused with real actual intelligence. It is artificial. As in NOT REAL. Like artificial flowers. They are not real. Neither is artificial intelligence. It is not real intelligence, AI. It is man-made folly. Have they made a phone tree work intelligently yet?
The silicon valley love affair with anything computechno seems to be driving the fantasy that computing will be a magic silver bullet. Which sells well. Very well.
They cannot do closed captioning worth a shat at IBM, the computer company, just watch the weather channel (they own) with cc on. Watch some of the sailing channel videos at utube (owned by Google), you will be astounded at the level of cc incompetence. Of what should be the most very basic AI work. Of course they can't make a car drive itself. What intelligence is calling something FSD 'fully self driving'? Ya gotta be kiddin' me! And what intel is buying it on that BS pipedream?
In fairness, they have not taught a lot of real people to drive yet either. But I can't believe how any people sacrificed themselves proving self-driving doesn't work. Or how successfully it was marketed and sold, and bought, all a mirage. Anything but do real concrete things that could actually work. Don't worry, AI will save the day.
All the money in the world for shat that is not real, fast as you can ask for it, see FTX or Theranos as Sima posted yesterday. For pipe dreams like carbon capture and cold fusion. Whilst nary a needle moves on homeless, hungry, fair min. wage or housing. If you really want to see some action, have a war.
The problem is the Klaus Schwab type Bond villians that really really want to be masters of the universe. They are in overdrive working on it right now. So obsessed with their brilliance and control craze as to have lost sight of reality. We have been and will be, their victims it seems. They think they can compute their way out of the mess they have made.
And like EL points out, the human factor will always poison it. Early reports on chatGPT and the new Bing AI sound like nightmares. People are asking to bring back Clippy and Victor it is so bad.
so glad I am on the back nine...
Hope it is all good out there for all!
We cannot solve our problems with the same thinking we used when we created them.
Look deep into nature, and then you will understand everything better.
both - Albert Einstein
Testify my brother...
righteous rant...as I stroll on up to the 19th hole.
Great wisdom, dystopian
What happened to sane and insightful thinking? Thinking the same misguided thoughts, as your sig line implies, will never benefit life. ‘Advanced' technology (thinking?) sure is causing havoc in the natural world.
I'm happy to share this certifiably sane place with you and others.
I couldn't help thinking
...of Edwin Black's book, IBM and the Holocaust. I see he notes in the introduction, that the holocaust would have happened anyway, but would it have been on such a scale?
He mentions "just in time" logistics that evidently were assisted by the punch card innovations being applied to deliveries to the death camps. Early in my history studies I noted that a few scholars felt that compiling lists of targeted groups was a feature of totalitarian regimes. The basic list in Edwin's Black's book is the census. I wondered about the concept of "human terrain mapping" undertaken in the US war in Afghanistan. Was this an anthropological project or another list or data base for computer driven solutions?
Scott Ritter notes the overwhelming importance of logistics or throughput as it was referred to in some military courses, in conventional military operations in Ukraine. If I recall correctly the final scene in Catch 22 is of a long line of tractor trailers at loading docks. Looked something what one sees at an Amazon distribution center every day. Of course, you must have quantity to put it in the trucks. I never understood how that worked economically, because the trailers at LTL carriers are mostly empty. They call it less than a truck load shipping. As trailer space is taken by LTL, routine shipping costs for bulk materials, essential goods, like food, hardware, construction supplies, etc., climb. So now when you go shopping for food or other mundane goods, there are items that are "not in stock," and the retailer is not sure when they will have the item, maybe in a few weeks. "Did you check online?"
Ironically, the Russians appear to have throughput and NATO doesn't. Maybe that's because they aren't as "advanced" as we are. I believe someone mentioned the utility of AI in solving "gaps" in the supply chain. I actually thought that there are some things that JIT can't fix, like the proverbial "smart war" designed by the current MIC. The MIC produces widget platforms that do everything in the "battlespace," because everything is "networked" to allegedly establish "full spectrum dominance." How does that work out in the conventional conflict in Ukraine? The joke of course has been for years, that the out of control military budget in the US would ultimately only be able to afford one missile, one ship, and one tank. Or perhaps they could only produce and deliver one of each. Then there is the overwhelming and expensive complexity of those widget platforms which makes them too expensive too difficult to operate in warfare with a "peer rival," whom our leaders treat with contempt.
What do I know? I have a humanities education. I've made my criticism of language AI before. I infer or guess things from the omissions and mistakes they make. So I can't say I don't use it. My primary use of translation applications is to get the script of traditional Chinese (Hanja) or Hangul rather than the meaning. Usually, I have to do it phonetically because AI knows nothing about usage or idiomatic speech. Phone trees drive me up the wall. No I won't do it online.
語必忠信 行必正直
When translating Classical Chinese texts
...directly into English, one had to discover the meaning of a character at the time it was written. A meaning not derived from translations into other languages, such as German, and a meaning not derived from missionary translations — even in the target language. This inevitably required creating a private database listing each translation of the same character throughout a document, and across all writings of the same period, allowing for context. As far as I know, there is no widget for that. The process altered the human mind, and only then would the correct meaning suddenly drop into place. Actually, it was the opposite of Artificial Intelligence.
Context is everything
According to my understanding that 60 percent or more of hangul was derived from Chinese roots or at least originally written in traditional Chinese characters. I'm hardly knowledgeable on Chinese itself. Sometimes I fool around reading Korean aphorisms as they are expressed in traditional Chinese characters, or try to interpret subtitles 자막 in Mandarin (those in traditional Chinese) because I wonder if it relates to the English translation. This requires that I pull out a four language traditional Chinese dictionary (1700 pages) which is really cumbersome and time consuming. I cannot read simplified at all. I can read the limited number of Hanja that you might find in Korean news stories for clarification of meaning.
Google AI, cannot translate the term "four letter idiom" from English to Chinese, from Korean to English, nor can it translate the Korean 사자성어 to Chinese correctly. This is actually a pretty common term.
http://www.koreanlii.or.kr/w/index.php/4-character_idiom?ckattempt=1
There are one hundred idioms translated to English at the link.
I also like Uncle Hanzi's database for Chinese etymology which I've posted before. Aha! Oracle bones.
https://hanziyuan.net/
Someone hit on one of my old posts translating the 1948/9 popular tune Rainy Gomoryeong, yesterday. I had forgotten it. This is part of a music historical project recording (a cover) by Chu Hyun-mi. I know it's not for everyone, but I love the lyrics.
Looking back to the time I let go mother's hand
The owl cried and I cried too,
I long for the evening we crossed over the mountain ridges
with fallen leaves flying about.
How many years was it, I picked the amaranth blooming
Behind the water mill, enraptured?
How could I ever forget, the forget-me-not's pledge?
When will the falling rain cross Gomo pass?
Tears of childhood life, how many hills were there?
Among lanterns closely hung in the Tavern house,
Writing these sentimental notes to myself,
This evening yet, the night sings the homesick song
작사 유호 Lyrics Ryu Ho
작곡 박시춘 Music Park Si-choon
노래 현인 Original artist Hyun Ryin
語必忠信 行必正直
sounds sad
yet sweet in almost
a Romanian (Gypsy) way
may be the low vocals?
thanks
question everything