Any sufficiently advanced technology is indistinguishable from magic”. - Arthur C. Clark Artificial Intelligence or AI is little more than advanced computer programming. First, Create an encoding schema to use bits to represent textual information that can be stored on a computer. Next, have a lexicon of words, grab “words” (substings separated by spaces) and compare against lexicon to identify potentially mis-spelled words. Next evolution, build a library of grammatical rules and read strings for adherence to those rules, and suggest updates based on those rules. (In the timeline of things we are here.) Maybe next will be to enable contextual flow of ideas. As I read through YD1BOR post on page 1, I thought that SDR already performs some of those functions. Some logging software already has the capabilities inferred by his idea of what “AI” can do. WY6K provides some great examples where AI may be useful by taking large data sources and using models (computer algorithms) to sift through the available data to provide a response based on the specificity of the query in a much faster time than we humans are capable. And that is the magic of AI, and why everything that people can’t distinguish from advanced tech is being labeled as AI.
A.I.? Artificial Intelligence. Hmmm....I've heard about it. Suspicious of it too. Is computer chess an example of artificial intelligence. Still got my Fidelity Chess Challenger and he still kicks my behind. Kraus/KC4ZGP
ChatGPT says yes, computer chess is AI, and I would agree. There isn't a clearly defined consensus of what things are AI or its specific meaning. NASA uses this from the National Defense Authorization Act of 2019 - https://www.nasa.gov/what-is-artificial-intelligence/. By that standard I think the automated railroad crossing gate near me is AI. I would definitely consider the system of road sensors and cameras used in Huntsville to regulate its traffic signals as AI. Do those "control" people? Sure. I mean, it is called "traffic control", and we are the traffic and pedestrians. Do I fear it? Not in the least, even if it can regularly beat me at chess.
Not really. Stockfish is just a mathematical model (as is AI, but stockfish is more rudimentary). There are some trying to use AI, but not sure about how much success it's had. Stockfish came long before most of the AI craze happened, so it doesn't use say open AIs models.
Stockfish apparently got a severe beat-down by Deepmind (AI). https://www.chess.com/terms/alphazero-chess-engine
There are other problems too. When the manual twist pictures together with words for to understand. Here the computer is out. The picture is a part of the manual for the tape machine REVOX B77, were the spurs of the tape are declared. Here AI will get problems for to understand fully in those pictures.
What we call AI today is little more than parsing “big data” and applying probabilistic responses based on predictive models (computer algorithms). At the heart of it, it is simply doing that which computers are designed to do; process complex data faster than the human mind. The outcome is entirely based on (and limited by) programmatic directives designed and developed by…humans. We are witness to the applications of Moore’s Law playing out in real time! But for a lot of folks it’s magic (hence the reference to Clark’s quote), so we give it a name (probably for commercialism and because it sounds cool). But we are not even close to what I would consider artificial intelligence. The impediments to unleashing the full power of computers will always be limited by our imagination of what is possible, and an inherent fear of losing our individual autonomy. Regardless, AI will always lack true emotion, unpredictability, and wisdom.
Just for the fun of it, I put that photo into blackbox.ai and this is what it gave back Magnetic Tape Recording: Understanding Track Configurations This diagram illustrates various track configurations used in magnetic tape recording, with labels in German, Dutch, and French. Explanation The diagram compares different track configurations, showcasing how the tracks are arranged on the tape and how they influence the recording process. Each track holds a separate audio signal, and the number of tracks determines the type of recording: 1. Stereo Recording Two Track Recording: The simplest form, where the audio is split into left and right channels, represented by "L" and "R" tracks. Half Track Recording: Similar to two track, but the tracks are recorded in half-width format. This offers greater capacity on one side of the tape. 2. Mono Recording Mono Track Recording: Uses a single track to record a single audio signal. Two Track (parallel recording): This involves using both tracks to record the same mono signal, increasing the audio signal strength, and potential use of a different recording medium. 3. Multi Track Recording Four Track Recording: This allows the recording of up to four distinct tracks, offering more flexibility for multi-tracking or recording in different modes (like quadraphonic). Quarter-Track Recording: Similar to four-track but the tracks are a quarter-width each, offering a higher density. Viertelspur (Quarter Track): This mode uses four tracks, numbered sequentially 1-4, to record multiple audio signals. Additional Notes: The diagram highlights the different starting and ending points of each track, crucial for synchronization and playback. The "REC: L/R" notation indicates that the recording head is capable of recording on both left and right tracks. The diagrams utilize arrows to represent the direction of tape movement. The Dutch, German, and French terms for the configurations are listed alongside their English translations. This diagram provides a visual guide to help understand the different ways audio signals are recorded on magnetic tape, offering valuable insight into the history and evolution of audio recording technology.
Thats terrific and an excellent use case. I had PDF versions of the various ARRL latest editions handbooks from when I bought the hard covers, and I just wanted to gather from multiple places within these extensive and detailed materials and handful of specific references and it did a brilliant job. One topic I wanted to double click on when I was cramming for the Extra exam and had it create a study guide and do a voiceover while I was running through other topics and it did a superb job. So many use cases given the LLM's are not off halucinating from unknown sources, it's just an LLM layer on strictly what you upload which is fantastic. For me doing installs/builds and tweaking antennas, I'll keep those books in, throw in a bunch of Palomar Engineers decks and other great sources and it's like having a personal HF coach
Not bad, I never thought that these infos came by correctly. AI got better understandig of this than much other humans. Often those figures, together with the real panel options, confuses.
I am reminded of Donald Fagens "IGY"... "Just machines will make big decisions Programmed by fellas with compassion and vision. We'll be clean when their work is done... We'll be eternally free and eternally Young." That was the vision back in 1958, the international geophysical year. And there's a subtle point nestled therein... For the machine to know what's going on, what's good, what's bad, what's fact, what's fiction, etc, someone has to teach the machine. Since computers especially large ones can make human mistakes with much greater speed and efficiency, the question must be raised, who do we trust to teach the machines, and who do we trust to monitor their output? And look by no means am I playing the Luddite here. I'm not suggesting at all that we should shove the technology up on the shelf someplace and ignore it. It's here. We're not going to be able to put the genie back in the bottle. But there are a large number of philosophical and moral questions to address as this thing grows. Those of you who are using grid tracker are already using AI of a sort when it predicts what the muf is going to be for a given area of the world. You're already using AI of a sort when it spell checks you as you're typing. In the end, just about any computer operation is some level of AI which in turn is little more than a series of if/then statements. I suppose we can argue about how much in the way of if /then constitutes what we now call AI. But in my view, AI has been coming for some time and growing. Some of you will remember that I have already mentioned that I run a heavily computerized shack. I find it a very effective way to operate. That said caution is advised. There's a limit to how far I plan on taking this.
This kind of freedom isnt a part of the normal social society and its rules. It steers into a Star-Trek generation and nobody knows until now how to handle this.
Hardly. Claude couldn't even get the specs right. Gave me a whole load of baloney about how it has a 235hp, 10 cylinder, diesel engine. Another time I just typed in "information for the Oliver model 237413 farm tractor" It came back with all sorts of made up information. There never was an Oliver tractor made with that number. My point: If AI is/was so good and accurate, it should have been able to tell me that those tractors were never manufactured. I don't think that I should have had to correct Claude. By having to correct and verify the information, I'm not saving any time.
