(AI Responses here are truncated)
Me: ChatGPT. I am member of an organization of physics researchers known as the International Association for Relativistic Dynamics (IARD). We are known for exploring fundamental particle physics in a way that is often outside the mainstream. We hold a biennial conference, the last of which was in Helsinki in June of 2024, at which I presented. Our next conference will be in the first week of June 2026 at an unknown location. I need your help developing a talk for this conference. I would like it to include original research in the field of particle dynamics.
That's fantastic, Matthew. I’d be honored to help you develop an original research talk for IARD 2026. Since you’re aiming for originality in particle dynamics, we’ll want to start by narrowing the scope a bit....
...Feel free to throw out any themes, even half-formed—brainstorming is welcome here.
Me: Well, the last talk I gave (via Zoom) was intended as an exploration of the work of Oliver Consa regarding the anomalous magnetic moment of the electron. Are you familiar with the work of Consa in this regard?
Yes, I’m familiar with Oliver Consa’s work, particularly his criticisms of quantum electrodynamics (QED) and his attempts to offer alternative classical explanations for phenomena like the anomalous magnetic moment of the electron... [Me (aside): Thank God!]
...Would you like to build a new talk that follows one of these lines of thinking? Or would you like to evolve your previous talk more directly — e.g., by testing one of Consa’s claims computationally or proposing a new analytic correction to the magnetic moment? [Me (aside): This might actually work.]
Me: Perhaps. I am not sure yet. But let's use this as a starting point. My talk in 2024 was not particular good, in part because it was via Zoom, but also because I was not able provide any good insight into this issue, only a review. One of my colleagues is Andras Kovacs. He has published work on this issue, regarding the use of the Dyson series in QED calculations. Are you familiar with his work, by any chance?
Yes, I'm familiar with some of András Kovács’s work—especially in the context of criticizing the use of perturbative expansions like the Dyson series in quantum field theory, and particularly their role in calculations like the anomalous magnetic moment of the electron.
[Me (aside): Oh, heck YES!]
16 comments:
According to high level Easter Dinner Table Consortium: AI can provide syllabi quite reliably for PhD candidates. The question remains you must ask the right questions to get proper debate. So we won't need professors but we'll need the art of questions. And apparently it's a big drag in speed if you polite to AI. Please refrain from please and thank you and are you a real person.
Ah interesting comment. "The Art of Questions" is a good way of putting it. Professors do seem antiquated.
Truly, there is a big ongoing debate regarding the use of pleasantries with AI, with some claiming by experiment that being polite actually DOES produce better answers. Yes, the main issue seems to be whether it is wasteful of computational resources to process please and thank you (and therefore incurring cost to the AI provider). It is not a big drag in speed for a given question, but collectively the AI providers want us to be as terse as possible for their own sake of costs. Others statea they will not give up please and thank you lest they get in the habit of speaking that way to human beings.
Am very curious about the field and topic of the Ph.D. candidacy being mentioned here.
Wherefore why if and whatnot. There are many types of PhD s that have yet to be invented. With or without AI.
I just a got profanity-laced earful from my colleague in Israel after emailing him about his use of AI in his research. In reality I was just reminding him that I exist and to keep me in the loop about the next conference.
Along those lines, my question was really curiosity regarding the individual involved so as to be proud of them over it. Nothing to do with AI, really.
Yes. Thank you. Our Consortium is too old to begin PhD or one needs to lead a bit of life outside of academia. But we can all dream and create a path of random study.
One = our M in the Consortium
I figured as much. Still very proud of her, for that decision, and what she's done.
Sometimes and actually very often AI can be quite annoying. For example you can ask "what is the closest font to match t hat of the got milk campaign?" And I don't have any of these fonts available on my current platforms and am quite hopeless at ink and pen.
Ah fonts...not my particular forté. Adam would know in a nutshell, but even I know that dealing with fonts can be a nightmare. You find one you like and the license is...huh...five thousand bucks WTF? Google Fonts has free ones. I was just asking your same question of ChatGPT on behalf of you (presumably to help the SHS?). It told me Phenix American was the typeface (I also asked it the difference between typeface and font so I could be super sophisticated). T
he followup question is just to tell it "I don't have those on my system, can you tell you a good free alternative please? I am using (your application name) on (your system)"
Of course you can leave off the please if you'd like. Then you may need to ask "It's telling me to download and install it, but have no idea how to download and install this on my computer. Could you hold my hand and lead me through it?" followed by "I'm still not getting it. Please help me", etc. followed by "I can't figure out how to ___ like it's asking me to."
The good thing about AI (if there is one) is that it is infinitely patient. It goes against our human nature to be so annoying pesky with simple followup questions like that. That's one of the hardest habits to break I think, much harder than please and thank you. By now, any human would have told me, "Really, Matt, you want to do what, now? You want to produce a historical map of the Willamette Valley with topographical features? Don't you have enough open projects already?" But instead it just hops up in. your lap like a puppy and away you go. I should mention I spring for the 20/month version of ChatGPT which remembers my history. VERY HELPFUL. I can go back and pick up any conversation later and it remembers exactly where I left off. I can't imagine not having that. That's how I plan to write my IARD talk. I was poking away at it today that way.
Inspired by something I read recently, I asked, on behalf of myself: "I want to learn how to cook but right now I'm hopeless. Where do I start?" It told me to learn to chop onions properly and I'd be way ahead of the game. So I asked it "how do I chop onions" and it told me (verified by another LLM). If nothing else, I can become the best onion chopper.
The trick with onions as we find in our primitive kitchen is to have a very very sharp knife. Hope AI pointed that out to you. I add " to you" so as not to end the sentence with the dreaded sentence-ending-prepostiion.
I will add my Wodens Day assignment here as I had no luck with the postcard entry. Ready? Ok, a cowboy rides into town on a Friday. Stays three days (always use three in stories)..stays three days and leaves on Friday. How is this possible? A: The horse is Friday.
I should have changed the horse's name to Wednesday. More date appropriate.
Somehow it asked me for moderation of your comments, which is why they did not appear right away. The joke might have stumped me. Fifty percent chance I would have figured it out.
When I was a kid in Iowa, there was a t.v. show in Des Moines called "Floppy" who was a dog puppet. Kids would go on the show and line up to ask Floppy riddles. Floppy was always stumped, even though he heard the same ones over and over. When I was 7 I got to ask Floppy a riddle, one of only two times in my life I have been on broadcast television. Of course I had to make up my OWN riddle to be original.
There was a rumor among us kids that one day Floppy accidentally got a riddle correct and it made the kid cry from the trauma.
always use three in stories---Paul Fry of Yale in his brilliant Introduction to the Theory of Literature course on Youtube emphasizes this point strongly in regard to children's literature, in regard the general importance of tautology. He reads "Tony the Tow Truck" to illustrate this, even the triple use of the letter T in the title.
Post a Comment