There’s been a lot about AI – not Amnesty International but artificial intelligence. There is concern that it will displace human beings: compose texts and music indistinguishable from what humans could. Much of what it spews out could be libel, propaganda, pseudo-science, and other misinformation.
There’s usually not much attempt to explain how artificial intelligence differs from the rest of computing. I think it is based on writing programs that learn from their own results. If equation 1 produces more fulfilled predictions than equation 2, drop equation 2. And it uses “large language”: enormous quantities of digitzed text that it can search and compare.
There is an entity (presumably consisting of humans) called OpenAI, and it has created ChatGPT, a digital robot that can answer questions.
My friend Larry Bohlayer, creator of the popular annual Moon Calendar, sent me an article (from Popular Mechanics, reviewing an article in the journal Ancient Mesoamerica) about something he knew I had taken an interest in, the complex calendar systems of the ancient Maya culture of central America. Among the many cycles in those systems is one of 819 days, which had long puzzled scholars. They have now realized that if 20 of these 819-day cycles are considered, making 45 years, there are correlations with the periods of the planets. (I remembered that 20 is the favorite factor used by the Maya in their several systems.)
Larry then made an experiment: he asked ChatGPT to evaluate the article he had sent me.
The robot answered with 324 words. It listed a website, an online tool, and three books about the Maya calendar, which it must have found in its large-language database. The paragraphs about these five items read like, and probably are, the authors’ or publishers’ blurbs about them. “This book provides a comprehensive overview,” etc. Essentially the bot told Larry to go and read the books himself and make up his own mind about the 819-day cycle.
One of the books was “Maya Cosmogenesis 2012: The True Meaning of the Maya Calendar End-Date” by John Major Jenkins. This, published in 1998, was based on the debunked idea that the end of one of the cycles, on 2012 Dec. 20, amounted to a prediction by the Mayans, when they set up their calendar around 200 AD, that the world would end on 2012 Dec. 20. After visiting the Maya temples, and making the cover painting for Astronomical Calendar 2010, I summed up the many ridiculous aspects of that scare story.
Larry then had fun asking the chatbot about me. It found bits of my own blurbs about my Astronomical Calendar; and I was delighted to find that on everything else I was safe from it. It said I was born in 1932, studied at the “University of London,” live in Connecticut (all wildly wide of the mark); it gave me credit for publications by other people, and magazine contributions I hadn’t made, and “several awards” that I wasn’t aware of receiving.
Judge for yourself the distance remaining between artificial intelligence and the brain.
__________
This weblog maintains its right to be about astronomy or anything under the sun.
The difference between AI and human intelligence is feelings (emotions), per Dennis Prager, talk show host.
He said this while explaining how men and women achieve maturity.
As girls mature, they need to learn to control their roller coaster emotions. They should learn to say “So what” more often. He gave the example of a woman who constantly nagged her husband for throwing his dirty clothes on the floor. Then when her husband died, she wished she could see his dirty clothes on the floor.
For boys to mature, they need to rein in their emotions of lust and anger. The natural man wants to spread his seed far and wide but the disciplined man learns to be loyal to only one woman. They also need to learn to say “So what” more often, rather than getting into fisticuffs over minor things.
I really did. Thought she had a sexy voice. Corrigenda: I actually got a lecture about pedology.
The dangers of AI were in 2001 A Space Odyssey.
Open the door, Hal.
I can’t do that Dave.
Several years ago I read about a DARPA study (Defense Advanced Research Projects Agency), or a study funded by DARPA, regarding the challenges associated with widespread adoption of AI by society. The study found that there were three significant impediments to widespread adoption of AI among the population: (1) people want to have physical money in their possession, (2) people want to own their own cars (or transportation more generally), and (3) people want medical attention from a human doctor whom they talk to in person. The study recommended that steps be taken to socialize among the population the benefits of giving up these desires.
(1) central bank digital currencies are being discussed now and will probably soon be implemented, which will remove the last bit of financial autonomy that people have now when we can still have dollar bills or coins in our pockets and pay for something without government knowledge or control. We saw examples of what governments will do when they have that kind of control during the trucker protest in Canada last summer ~ they will simply freeze bank accounts whenever citizens do something the government doesn’t agree with. Your money is not your money unless you never deviate from the government defined limits of behavior or speech.
(2) The drive for electric vehicles which are connected to the internet, and which already comprise components that you don’t own even if you “purchase” the car, is part of the movement to establish the conditions for centralized control of transportation ~ the authorities will have the ability to limit or completely restrict the movement of such cars at a moments notice if they choose. They will also push for the adoption of the idea of cars as ubiquitous “public utilities” that you can access as needed but simply leave in place once you’re done with it. Example: right after the Russian invasion of Ukraine, people floated the idea that Elon Musk should “turn off” all Teslas in Russia. As crazy as the idea sounds to me, worse is the reality that he could have done it if he wanted!
(3) During the COVID pandemic we certainly saw the widespread adoption, throughout the western world at least, of the idea of centralized “standards of care” ~ with the CDC or NIH here in the U.S. defining exactly what was and was not allowed in terms of treating COVID, which although I’m not a doctor, I have read entirely contradicts the ideal of a doctor-patient relationship in which the doctor has license to develop a treatment or care plan based on his or her individual diagnosis of the patient and based on their own knowledge and even intuition as a professional. There certainly were cases where doctors recommended treatments for COVID that were not aligned with CDC mandates and they were fired or threatened with losing their license to practice medicine, despite being effective treatments in many cases. I believe central authorities who want to control their populations better would love to go beyond “standards of care” and have most people interact with an AI enabled computer for their medical needs.
As far as ChatGPT is concerned, it’s an amazing tool to write software ~ you ask it to help you write the html code for a web page that does something a little complicated, and it spits out the code in a matter of seconds, and then explains to you what each section does and how you need to adapt the code for your own data. Very cool!
I asked Alexa to talk dirty to me and got a lecture about geology.
Did you really or is that a good joke?
In Joyce Cary’s gorgeous novel “The Horse’s Mouth”, one of Gulley Jimson’s ruses to scrounge a living is selling “dirty postcards” – plain postcards that he has smeared with dirt. As usual it ends badly for him.
In natural language, words endlessly develop shifted senses, and AI will have to try to keep up with these.
They are working to exert FURTHER their mastery over their people(s). Population control requires individual exertions (or non-exertions!).
Both!
The topic worries me a great deal. You can bet that right now, there are governments around the world working on ways to use AI to control both their and the world’s population.
John, by “control” do you mean they are working to exeert mastery over their people or to control their population size?