Welcome to the StarCast for the week of June 29th, 2025. I'm your host, Jay Schaefer, and with me is co-host Mike Lewinsky. Let's take a look at space weather from SpaceWeather.com. Mike, what's happening with the sun over the next couple days?

[Mike L.] Well, there are some sunspots out on the disk of the sun facing the Earth, but they are all stable and quiet. There are no solar flares expected for the next 24 hours. And the possibility of geomagnetic storms is very low. We get to maybe a 20% chance of active conditions in the next day. And it the conditions are only smaller, down to a 1% chance of a severe storm, so we're not expecting to see any northern lights here at Mid-Latitudes.

[Jay Shaffer] Oh, okay. Well, that means I can stop aiming my camera north and aim it south to catch the beautiful Milky Way now. So tonight, if you look west, to the west, at the crescent moon, you should be able to see a little red dot that is Mars, and they're very close together, only about a degree apart, depending where you are in the U.S. And so, if you want to see Mars and the Moon together, kind of look up toward the west, and find the crescent moon around 9 tonight, and you should be able to pick out Mars as well. And with the moon being just a sliver, this is one of the best times that you can see the Milky Way. And then, of course, you know, if you want to get to a non-light-polluted area. This is the time of the year where you can see the stretch of the Milky Way that includes the Cygnus Starcloud, which is one of the most star-rich regions in the night sky. And when we look towards Cygnus, we're looking downstream through the local arm of our galaxy, which is called the Orion arm of the galaxy, and we're looking toward the core of a galaxy, toward the actual black hole that is in the center of the galaxy in the constellation Sagittarius. So those are kind of the main events in the night sky this week. There was a couple of pretty interesting events in the news this week. First, there was the images released from the new Rubin Observatory, and then there was the fireball meteor with meteorites in the southeastern United States. On June 26, approximately 12:25 p.m. Eastern Daylight Time. A bright daytime fireball streaked across the sky, culminating in a thunderous explosion Southeast of Atlanta, Georgia. The American Meteor Society, or AMS, was flooded with over 200 reports from 20 different states as people witnessed this dazzling object moving from North, northeast to south-southwest. Mike Hanke, the AMS Operations Manager, noted that many instruments recorded it in the fall, including the National Ocean and National Oceanic and Atmospheric Administration satellites. And Doppler radars, and even some of our all-Sky 7 cameras. Shortly after the sonic boom that many people heard. A resident in McDonough, Georgia, about 30 miles south of Atlanta, had a truly out-of-this-world experience, a golf ball-sized rock punched right through the roof, through the ceiling, and landed on their floor. And thankfully, nobody was hurt. And this quickly drew the attention of meteorite hunters from around the U.S. Who were on the lookout for what they sometimes call charcoal briquettes. This term refers to the fresh black and thin fusion crust that forms over the newly fallen meteorites during their fiery dispense through the atmosphere. And if they weren't disappointed, Stephen Dixie of Atlanta was on the scene the very same day, and successfully recovered two beautiful stony meteorites from this incredible event. So, yeah, and so, you know, this was a meteorite Hunters dream day, and of course. A lot of people in the research community would rather that meteorite hunters would turn the meteorites over to the universities, but there is some money to be made from hunting meteorites. So, Mike, what's happening with the new Rubin Observatory?

[Mike L.] Yeah, Jay, the Rubin Observatory released their first batch of images during an online livestream event this past Monday. Images from the new telescope included the Trifid and Lagoon Nebulas and the Virgo cluster of galaxies. The observatory also announced discovery of about 2,100 asteroids, including 7 near-Earth asteroids never seen before. None of those pose any risk to the Earth. But we're it's gonna continue to catalog new asteroids. So, these images in the June 23rd First Look event are just the beginning. Every three nights for the next decade, the fast-moving and highly sensitive telescope will complete a scan of the southern hemisphere sky, generating 60 million gigabytes of data. And processing that much data is going to require the use of AI. And we'll talk a little bit more about AI and astronomy in the next segment. But the telescope is named for the astronomer Vera Rubin, who is well known for discovering firm evidence for dark matter. And the observatory is going to generate survey images consisting of exposures of 15 or 30 seconds, using 6 filters to cover the full visible range, as well as near-infrared wavelengths. So, when these events hit the media, they're often sensationalized, and as they filter through social media, they become even more sensationalized and oftentimes deceptive. So this week, our discussion of the podcast is Junk AI Images, misinformation, and exaggeration about astro events.

[Jay Shaffer] Yeah, and I think this is particularly aggravating for more seasoned observers that this information, you know, gets eaten up by novices who don't really can't really discern, you know, some science facts from some of these kind of exaggerations, and and deceptions.

[Mike L.] Yeah, and it's become its own mini-economy. You know. The AI slop, as it's called in general, and you know, with space news and astronomy photos in particular. People have learned that they can. Post-sensationalized and exaggerated photos and generate a lot of engagement and start to monetize their following. And so. The incentives are upside down here. The real images are not going to compete with the most highly creative generative AI that can create things that simply don't exist. And so, those are the things that attract the most attention, the most likes, the most comments. And it is a very, like I say, upside-down incentives that favor the fake and the sensational over the real, and we start to lose our appreciation for what's real.

[Jay Shaffer] Yeah, I'm and I've noticed, particularly on social media, that, you know, a lot of these accounts, you can, you know, if you kind of keep a close eye on, you know, what the account is, you know, so there's an account, for example, that's called Our Amazing Universe, and there's no name attributed to it, and it spews out, you know, bunches of AI images and AI headlines, and for example, you know, here's a couple of 3 or 4 from last week that I noticed on social media. So there was a a account called End Times News, and and had a big AI image with hundreds of meteors, you know, falling with the Milky Way in the background, and and, you know, kind of the sunset image, you know, where, you know, there was kind of red glow on the horizon, and the headline was, booted Meteors, Pelt Alaska, you know, as shown in this image. And so some of the other ones that I saw, you know, one relating to the then Fireball, the Georgia Fireball, was meteor burning over metro Atlanta, and it showed, you know, the skyline of Atlanta with a, you know, a world-ending-sized meteor image on there. And then another one was fireball slams into Georgia home. And so, you know. And then. And then, relating to it, there's a there was a comet that's way out, still, I mean, it's a very large comet, it's what we commonly call it Comet BB, and it's still way out beyond Neptune. But, you know, the scientists were seeing some exploding jets coming from this comet. And so, the sensational headline was, Mega Exploding Comet Headed Towards Sun. And so and the exaggeration, of course, is this comet won't even, you know, come within, like, 9 AUs of the Earth, and although it is, you know, huge, it's just and it's not gonna get close to the sun until 2031. You know, but they had to make this, you know. Big this is gonna end the Earth type of headline. So what are you seeing recently, or what are some of the more memorable deceptions that you've heard about or seen.

[Mike L.] Yeah, of course, we don't need AI to create deception, and you know, I think one of the most classic and memorable composite images. That was clearly fake, showed a very thin crescent moon. And there were stars inside of the crescent, so somebody had taken that, you know, with a mask, and made a mask of the crescent, and then popped it into a star-filled sky with just zero understanding of what the, you know, actual appearance of a crescent moon is, and what the structure is. I also saw an image that had, it was a Star Trail photo, so either a long exposure or a stack of long exposures. And the Star Trails in the water of the lake that were reflected. We're all distorted by the currents. But they were distorted in a way that, you know, the person who had made the composite or made the AI slop, you know, did not understand that this was a long exposure. And so, the ripples of the currents had nice curved star trails that, you know, would curve back and forth in a snake-like pattern. That simply could not happen. It was a you'd taken the long exposure of 4 hours, and then, assumed that that was produced instantaneously. And that the reflected image and the ripples in the water would perfectly distort all of the star trails, so. Sometimes it's just really obvious that an image has been altered, and sometimes less so, and I find, you know. My skepticism is so heightened these days that I'm reluctant to engage with anything that looks too perfect, and I know that that's that some of what I see that's too perfect is just the result of very talented astrophotographers doing what they do best. But I don't want to inadvertently bolster visibility by interacting, liking, commenting. I mean, you know, I think the worst thing that we can do with an obvious AI slop is to comment on it, because that's the interaction that they're trying to drive. Even a negative comment of, this is fake. Puts it in front of more people's eyes. And so, by trying to debunk, we inadvertently monetize their product for them. And so, I've kind of switched to a whitelist approach of. These are reputable astrophotographers who I have followed. For many, many years. And I see them complaining about their work being used as the basis for generative AI, and often. They will post side-by-side, this is my photo from such and such a date and such and such a place. And then the generative AI that has ripped off their work without attribution, which is so clearly by the basis of landscape features and even much of the much of the image. In some cases, you know, they took what was an amazing image and made it over the top amazing. And looking at the AI slap, it's not clear. In a standalone context, not at all clear that. This is fake. But I do, you know, give more credence to those photographers who are seeking out and finding or having reported to them. Derivative rip-off. Slop. And so those are the astrophotographers I'd preferentially follow, and and boost, you know.

[Jay Shaffer] Yeah, and I think there's a legitimacy to where legitimate astrophotographers like you and I, I would venture to say, where we annotate our, you know, photos with, you know, the context of where it was taken, and what direction we were facing, you know, whatever our camera settings were. And even a lot of these astrophotographers who do composite images, and, you know, there's good reason to do some of this compositing, where they will actually, you know. Annotate in their post with the image to say, you know. This was, you know, this was a stack of, you know, 15, Milky Way images, and then I added the foreground, you know, with a different exposure, but taken on the same night from the same location. Or, you know, or, you know, they may even say, you know, I. You know, composited this this was a composite image of the full moon that I had to, you know. Take the exposure of the full moon down, you know, quite a bit to composite it into so that you could see stars in the background behind a full moon or something like that. And as long as they're kind of honest about that, and, you know, forthcoming. I think that that is good for the community and good for education, and good for novices wanting to, you know, be excited about the universe. But I think on the other hand is when we see these AI slop and generative AI images. They raise these expectations beyond realistic levels for novices, and so. And it leads to, a lot of times of them being either, you know, dismissive of real images, or underwhelmed, and therefore lose interest in, you know, the night sky and astronomy. And what's been your experience with that, Mike?

[Mike L.] Yeah, I think even even our, you know, traditional craft of astrophotography poses that risk of, you know, the camera sees so much more than the naked eye, and if people are accustomed to looking at, you know, legitimate, real photos of the universe. The experience of going outside at night, you know, and thinking, oh, well, I live in a city, I just can't see that, it's too bright here, but then you get out into a dark sky area. And you still do not see the level of detail, and certainly almost none of the color that is apparent in a photograph, and so. I periodically make the effort to post, you know, a side-by-side, this is what my naked eye can see. And this is what my camera can see, and it's a difference of, you know, a few milliseconds of photons entering my eye versus, you know, 15 to 30 seconds of photons entering the camera sensor. And so. The technology gives us the ability to appreciate. The universe in a new way. But we I think we do have a responsibility to periodically remind our audience that. The camera is so much more sensitive that you're not going to see this, and that's not a reason to not look up. It's it's, you know, get excited to go out and do the viewing and have reasonable expectations of what's possible.

[Jay Shaffer] Yeah, and you know, for example, when we look at the Rubin Observatory images, you know, I was looking at the Trifert and the Lagoon Nebula images that were released from the Rubin Observatory, and, you know, they were just brilliant, you know, with, you know, all this detail and the nebular clouds, and, you know. A wide variety of colors, and what a lot of people don't realize is, number one, you know, it was a longer exposure, and probably a composite exposure where, you know. Multiple exposures of, you know, 15 to 30 seconds. But they were also using… when we talked about those 6 filters, earlier, when we were talking about the telescope. Is those filters are, you know, bringing out the best. Light of, you know, the best filtering out the strongest photons in a particular. Frequency, which may or may not be in our visible light spectrum for our human eyes, and so. And so a lot of those images that you see of Nebula and deep space objects will have. You know, stuff that our eyes could never see, you know, with a lot of infrared, material, and that is basically compressed into. Our visual scale.

[Mike L.] That's right. And, you know, on this topic of how much information or how much, you know, light there is out there. I think it's important to make a distinction for the truly beneficial uses of AI in astronomy. There's an article last May in the Mit Technology Review, talking about the square kilometer Array Observatory. They are placing hundreds of thousands of dishes and antennas in deserts across Australia and South Africa. That's going to be generating some 300 petabytes of data a year. And one of the computational astrophysicists, quoted in this article, Cecilia Garofo, says, when you have too much data and you don't have the technology to process it. It's like having no data. And as I was, you know, thinking about our topic today and doing a little reading, I recalled Vannevar Bush's Paradox. And Vannevar Bush was instrumental in the creation of the National Science Foundation, which was established by an actor of Congress in 1950. But in the years preceding that, he had an article in the Atlantic. Where he lamented that specialization was causing a proliferation of information. He says there's a growing mountain of research. But we are bogged down, and we can no longer grasp, much less remember, all of these observations and conclusions. And so, the paradox here is that as information proliferates, our ability to access relevant and accurate information diminishes in proportion. And of course. He was worried about this in 1945. The article is excellent. It describes a device that sounds very much like a web browser when you read his description, it was, you know, visionary, and he. You definitely anticipated a lot of our modern technology. As needed, as the only way that we would be able to manage this mass of data. And, you know, I think of what despair he might have when he could see that this, you know, has resulted in. Not just, you know, staggering volume of scientific data, but an even greater volume of misinformation. Generated by people who. Just want to make a quick buck, and who, you know, can use generative AI to create thousands and thousands and thousands of permutations of news articles and images. In the hopes that one or two of them. Catch on and become viral, and they don't need to. Tune anything. They're just brute-forcing reality, as it were. By saturating the information environment with so much. That sooner or later, they're going to hit something that happens to catch people's attention. I've heard it said, well, AI is going to save us from AI, we're gonna. Develop new tools to sort through all this. Slop. Um, but.

[Jay Shaffer] Mm-hmm.

[Mike L.] We really seem to be locked into what the biologists call the Red Queen's race, where. Every innovation by the scammers requires more technology to filter it out, and then. In turn, the scammers adjust their algorithms and add more resources. That, in turn, require us to adjust our defenses, and. The end result is that we're locked in a perpetual arms race, where there are no winners.

[Jay Shaffer] Yeah, I was. I don't mean to interrupt you, but it was just, I was just discussing this last night about anti-plagiarism. Software for when I was, you know, teaching at college, is that was, you know, just when AI was starting to really become a. A thing, and of course, you know, and this became an arms race between, the, you know, the faculty. And the students, between their AI-generated, you know. Papers and, and, and our anti-plagiarism, and so. You know, when you're talking about that kind of arms race, that's the exact same kind of thing that's going on in academia with, you know, just, you know, writing, you know, basic. Papers and, and students. Oh, no. But, you know, and I agree with you that there's. You know, there's a lot of use for AI within astronomy, and, you know, we're not we're not talking about. Dissing AI in general, and, you know, like, you know, I was reading an article about. An AI framework that's called, the SimBig. And it's a simulation, inference of galaxies, and it's actually. Is trying to. Simulate entire universes, and so they basically will the AI model will, you know. Generate the simulated universes from the what it's learned about, you know, cosmological parameters, and then it basically. It uses some of our noise from our actual observed. Universe, and applies it to these artificial universes, and so this is. Pushing forward cosmological, research. And so. So AM.

[Mike L.] Yeah, that's right. It's it's all per they're purpose-built tools. This isn't just, we're gonna throw some papers into chat GPT and have it solve, you know, the mystery of.

[Jay Shaffer] Yeah, go ahead.

[Mike L.] Dark energy, by any means, but. But, utilizing, you know, astronomical. Computational science, and writing specialized applications to process data. We are getting benefits from so-called AI.

[Jay Shaffer] Yeah, and and also, one of the things that I. Need to talk about is just that even mainstream media is that. And I started my career early in my career was in broadcasting, and there was always, you know, at the end of a newscast, there's what we call the fluff segment. And this is basically. If they need to make, use some time at the end of a newscast, and they don't have a story to fill it, and they need to stretch it just a little bit. There's these rip and read, you know, basically news stories, you know, and they're always, you know. And. And. Of course, you know, astronomy. And scientists will put out these news releases, but they'll be kind of. You know, written in such a way that they could be, you know, misinterpreted or sensationalized, and so. And so sometimes, you know, like, I'll be. Turning between news channels, and I'll see the same story, you know. On an almost verbatim word by word, where these newscasters. You know, what we call rip and read, these, you know, press releases, and, you know, and I remember one back, I think it was about 2016. And it was somehow it got sensationalized that there was going to be a. Super Mars, and the Mars was going to be bigger and brighter than the Moon. And that's. And, and that got picked up by media. And. And the way that the, you know, kind of the original press release was, you know, it was a crescent moon, and, you know, definitely the dimmest that a moon could possibly be, so and it was also the. Apogee of Mars, which made as close to us as it possibly could be. And as large as it could possibly be, but of course, it was. Much, much, much, much smaller, you know, than the Moon would have been. And, you know, and I had actually people coming up to me, you know, are you getting your telescope out tonight to see if the moon, you know, see if the Mars as big as the moon? Can we look? And, so yeah, so, you know, sometimes you'll see these stories in mainstream media as well, and so, and.

[Mike L.] Yeah.

[Jay Shaffer] You know, and they and I don't think they intentionally exaggerate. I think it's just more of that they're not knowledgeable enough to discern whether. You know, they are exaggerating.

[Mike L.] I think you're right, and it certainly points to a need for higher quality, better educated, more science journalism.

[Jay Shaffer] Yeah, and. You know, as far as the mainstream media goes, or your local, you know. Your local news goes, is that. That often the only scientists in the room is the meteorologist, and so they're the person that. Basically, you know, talks about. Space and astro events, and. And, you know, that's a great thing, and, you know, and I think that, you know, meteorologists are. Doing a great job, but it's, again, you know, people kind of just look at it the same way that they look at the, you know, at the weather, you know, forecast. They might be right 50% of the time, you know?

[Mike L.] Right?

[Jay Shaffer] So when we, you know, I know. For you've pretty much. Are weaning yourself away from social media altogether, and I think I respect you for that a great deal. I'm still. Unfortunately, pretty addicted to, kind of, you know, Facebook, Instagram, and. And, you know, and of course, we're both on Blue Sky and that sort of thing. So, what do you think the, the kind of, solution, or what can, some people that actually do to kind of. Help themselves, discern what is real.

[Mike L.] You know, that's a great question, and there are not really simple answers to it. I mentioned earlier that I've kind of whitelisted a certain set of. Astrophotographers that I have been following, and landscape photographers and other photographers. I've been following for years now. And whose work I trust, who have. Too much investment in their craft and their credibility and their art. To use generative AI. Of course, you know, many of us do use. Tools for denoising that are based on AI in some way. It's not to say that there's zero use of it, but it's not. To create something that is unreal. Or, didn't happen. So, but for somebody who doesn't already have that whitelist built. It can be really difficult. I think that. Sticking with, reputable. Organizations that, you know. Are doing the vetting for you, then that could be from. You know, International Dark Sky. It could be from, university astronomy. Programs. Certainly, you know, getting out there, as we talked about last week, getting out to a star party. And just looking through telescope yourself. So you start to build up some real-world experience. I was in. Mesa Verde National Park this past week, and. There was a presentation given in the amphitheater at the. Campground, titled, Ancestral Skies. And at the conclusion of that presentation, we. Went up into the parking lot, and they had a couple. Telescopes set up, and people took turns. You know, looking through the telescopes with the Rangers giving a bit of a tour using laser pointer.
[Mike L.] You know, we're looking over there. At, you know. Hercules, and here's the cluster. And. So, I think getting some field experience and not completely relying anymore on. What we see in social media, just being. You know, skeptical. I think you can, you know. Start to discern the pages that don't. Clear attribution of authorship of photographers. Who are consistently posting these absolutely unreal images when every post is just over the top. You've got somebody whose farming engagement. Who's just there to make money, and they don't care about truth, they don't care about presenting reality as it is, they only care about engagement. So, block them. Block and report. Don't interact, don't like, don't dislike. Don't laugh or poke fun at, you know, just. Getaway. That's my best.

[Jay Shaffer] Or block, yeah. I block, but and then, also, like, I'm.

[Mike L.] Yeah.

[Jay Shaffer] A moderator, or, you know, I guess a resident expert in some of these groups, astronomy groups on. On Facebook, like, for example, we have a group called, Taos Skywatchers. And I remember when you were down here, you used to be a contributor and moderator there, and basically a resident skeptic.

[Mike L.] Mm-hmm.

[Jay Shaffer] And so I and so I kind of monitor some of these, you know, groups that are local. You know, people are seeing things like weather balloons or, you know, or a meteor, or a re-entry, and going, you know, what is this? And, and. And, you know, kind of, like, chiming into the group and saying, well, you know. If we've got a time, and we've got a, you know, vector where you were looking, and and, you know, and a good photo of it, we can, you know, we can look at, like, satellite databases and say, you know, that was a satellite. Or that was a satellite reentry, or that was a rocket launch, or, you know, and I don't know how many times. People who've seen Starlink trains for the first time with their naked eye, and going. You know, oh my god, alien invasion, you know? And, you know, and being on some of these groups, and being able to, you know, kind of be the resident skeptic, or as I call myself, the. Expectation, over-expectation manager for some of these groups. But, you know, so that's kind of some of the things that I'm doing. And of course, you know, that brings us back to this podcast. And so, you know, kind of, it's. Reason d'etre is to show that we can, you know, help. Novice… novices navigate, you know, what the information about our universe.

[Mike L.] Yeah, this is so desperately needed.

[Jay Shaffer] Okay, so why don't we go ahead and wrap up today, and we want to thank all of our listeners for checking out the podcast. Please be sure to comment, like, subscribe, and let us know what you'd like to hear more about, or less about. And you can also check out our individual websites. Mike's is wilderness Vagabonds.com. And mine is Skylapser.com. Our intro music is fanfare for space by Kevin McLeod from the YouTube Audio Library. From the Deep Space Nine Observatory, this is Jay Schaefer, and.

[Mike L.] This is Mike Lewinsky.

[Jay Shaffer] Wishing you all clear skies.