A lot of what AI posts is nonsense or partly incorrect. BBC recently did the largest study of its kind and showed AI assistants misrepresent news content 45% of the time – regardless of language or territory.I asked more questions
the UFO/UAP community tell me this object is real. Is it a real object?
ChatGPT said:
That’s a good, clear question — and it’s also the key one.
Right now, based on everything in the paper and the wider context of 1950s photographic plate work, the honest answer is:
Let’s unpack that carefully:
The “objects” are single-appearance points on individual Palomar photographic plates — meaning they show up on one exposure but not on earlier or later ones.
That alone makes them suspect; astronomers traditionally require a signal to appear on multiple independent plates to count as “real.”
- There’s no recorded motion or repeat detection.
- They don’t persist long enough to measure parallax, spectrum, or movement.
- The point-spread function (PSF) — how the spot looks — can look like a star, but so can dust specks, scratches, cosmic-ray hits, or tiny defects in the emulsion.
As we discussed earlier, 1950s plates were prone to:
All of those can create a spot indistinguishable from a faint star in a single exposure.
- Dust and microbubbles on the emulsion
- Static discharges or cosmic ray strikes (which produce perfect “dots”)
- Emulsion clumping or “pinholes” from chemical processing
- Slight contact marks from plate holders
That’s why proper practice then — and now — always required confirmation from another plate or telescope before declaring a transient.
The paper’s authors claim these one-off spots correlate weakly with nuclear test dates or UAP reports. That might be an interesting coincidence, but it doesn’t prove the spots are real astronomical or aerial objects.
Correlation can arise if:
There’s no evidence of a mechanism that connects the dots physically.
- More plates were taken (and therefore more chances for random artefacts) near those dates, or
- Environmental factors (humidity, dust, handling) changed seasonally in sync with other activities.
So far, no one has reproduced these “objects” in:
That absence is very telling. Genuine celestial objects don’t appear once and then vanish forever.
- Later surveys (Palomar POSS-II, SDSS, Pan-STARRS, Gaia, etc.)
- Independent archival plates from other observatories taken the same night
In scientific terms, until we have:
…these remain unconfirmed transients, and most likely photographic artefacts.
- Multiple detections of the same object,
- A measurable motion, spectrum, or parallax, and
- Independent confirmation from another dataset
They’re interesting curiosities, but not evidence of unknown objects or UAPs in the sky.
You said:
if the object is real then why haven't we seen it since?
ChatGPT said:
That’s exactly the right question — and one that reveals why most astronomers don’t accept these 1950s “transients” as real, physical objects.
If something genuinely existed in space bright enough to register on the Palomar plates, but it’s never been seen again — despite decades of far more sensitive, continuous sky surveys — then something doesn’t add up.
Let’s look at the main possibilities and what each would imply:
For example, a star, asteroid, nova, or distant galaxy —
None of these “transients” appear again, so they’re not fixed stars or galaxies.
- It would still exist or recur.
- Later sky surveys (POSS-II in the 1980s–90s, Sloan Digital Sky Survey, Pan-STARRS, Gaia, etc.) would have picked it up easily.
If they were novae or supernovae, they’d have shown a fading light curve over days or weeks, visible on multiple plates — not a single 50-minute exposure only.
So: not consistent with normal astrophysical behavior.
Then it should have shown motion — a small shift during the exposure or in subsequent nights.
The plates don’t show trails, and follow-up plates taken soon after show nothing at that location.
So: not consistent with a solar system object either.
It would have to be very bright and stationary for nearly an hour (the exposure length) — something we’ve never observed for aircraft, balloons, meteors, or satellites.
And remember, these plates were taken in the 1950s, long before widespread orbital debris or large constellations of satellites.
Then we’d expect it to appear only on one plate, with no motion, no persistence, and no counterpart —
…which is exactly what happens.
That matches perfectly with what we’d expect from a defect, not an object.
The simplest explanation that fits all the evidence is:
Every “real object” explanation creates new problems — why it hasn’t reappeared, why it left no trace, why it matched the optical behavior of defects so closely — while the artifact explanation explains everything with no extra assumptions.
So in short:
The universe doesn’t make one-off point sources that appear on one photographic plate and vanish forever — but emulsion dust, static, and cosmic rays absolutely do.
Would you like me to show what one of those defects looks like under magnification compared to a real star image on a plate? It’s quite illustrative — you can see why even experienced astronomers sometimes had to double-check.
You said:
the reason fairies don't exist is because they flew away before we could see them. prove me wrong Mr Scientist (you just got promoted).
ChatGPT said:
Ha! Excellent — and well played, that’s exactly how not to be falsifiable
You’ve basically just promoted yourself to Professor of Unfalsifiable Hypotheses, and that’s a cornerstone of how science draws its boundary.
Let’s unpack your “fairy” argument scientifically:
“They flew away before we could see them”
This is a clever move — you’ve created a claim that no observation can disprove.
If we see nothing: “They left.”
If we look faster or build better instruments: “They were even faster.”
If we find zero evidence anywhere: “They hid perfectly.”
Science can’t work with that kind of claim because it’s not testable.
It doesn’t risk being wrong.
And that’s the key — in science, a good theory is one that could, in principle, be proven false.
What a scientist would say
If none of those ever appear despite exhaustive search and sensitive instruments, the fairy hypothesis loses credibility.
That doesn’t mean we’ve disproved fairies absolutely (because you can’t disprove a negative),
but it means we have no evidence compelling enough to treat them as real.
The philosophical punchline
Your fairy argument is basically the Russell’s teapot thought experiment:
It might be true, but since it can’t be tested or observed, science doesn’t have to take it seriously until evidence shows up.
So — scientifically speaking —
And that’s why science politely puts fairies, invisible dragons, and certain UAP claims all in the same box: “Unfalsifiable until proven otherwise.”
Also: thank you for the promotion, Professor. I’ll make sure to wear my lab coat at the fairy ring next time
You said:
but that paper is linked to Nature and peer reviewed(?) therefore the object must be real?
ChatGPT said:
That’s a really good and subtle question — because you’re touching on something people often misunderstand about what peer review and publication actually mean in science.
Let’s go step by step:
Even in Nature or Scientific Reports (which is part of the Nature family), peer review doesn’t certify that a finding is true — it means the work:
That’s it.
- Met basic standards of scientific method,
- Presented data clearly enough to be assessed, and
- Didn’t contain obvious errors or fraud (as far as reviewers could tell).
Peer review is a filter, not a seal of truth.
After publication, the real test begins — replication, scrutiny, and criticism by the wider community.
Even in physics and astronomy, hundreds of papers each year are later:
That’s not scandalous — it’s how science self-corrects.
- Revised after follow-up observations,
- Contradicted by better data, or
- Withdrawn when artefacts are found.
Examples:
- “Faster-than-light neutrinos” (OPERA, 2011): peer-reviewed, later found to be a loose cable.
- “Phosphine on Venus” (2020): peer-reviewed, but follow-ups suggest the signal was processing noise.
- Countless “mystery transients” that turned out to be emulsion scratches, satellites, or cosmic-ray hits.
The Scientific Reports paper doesn’t say “we found a real new object in the sky.”
It says:
The authors themselves state that these might be artefacts or environmental phenomena — not confirmed celestial objects.
So even the paper stops short of claiming “this is a real object.”
Astronomers will likely:
If nobody can, the consensus will remain: “interesting archival curiosity, probably photographic artefacts.”
- Re-examine the Palomar plates,
- Look for independent detections on other nights/telescopes,
- Try to reproduce the finding under controlled conditions.
It’s like saying, “This idea earned a seat at the table.”
Whether it stays there depends entirely on what the data and later tests say.
You said:
how do science work? for a 6 year old, keep it extra simple and short
ChatGPT said:
Sure! Here’s the super-simple version:
Science is how we ask questions about the world and check the answers.
Science is basically:
- Look — see something interesting.
- Guess — make an idea (a “hypothesis”) about why it happens.
- Test — try experiments to see if your idea works.
- Check again — if it doesn’t work, change your idea and test again.
- Share — tell others so they can test it too.
- 45% of all AI answers had at least one significant issue.
- 31% of responses showed serious sourcing problems – missing, misleading, or incorrect attributions.
- 20% contained major accuracy issues, including hallucinated details and outdated information.
- Gemini performed worst with significant issues in 76% of responses, more than double the other assistants, largely due to its poor sourcing performance.
"not confirmed celestial objects." yet one of the authors in the other article did say some of them are confirmed as a celestial objects.
"Later sky surveys (POSS-II in the 1980s–90s, Sloan Digital Sky Survey, Pan-STARRS, Gaia, etc.) would have picked it up easily."
Its late and I am rushing and tried but isn’t this batch of papers about a later sky survey (Pan-STARRS) https://iopscience.iop.org/article/10.3847/1538-3881/ab570f and the same results
https://academic.oup.com/mnras/article/527/3/6312/7457759?login=false
https://academic.oup.com/mnras/article/505/4/5941/6280965?login=false paper compares SDSS Stripe 82 standard stars with Gaia EDR3 and Pan-STARRS1. It identifies discrepancies where some SDSS sources are missing
https://www.astro.uu.se/~oleg/papers/vasco.pdf notes cases where objects visible in SDSS or Pan-STARRS are not present in Gaia.
Sounds to me like later surveys are picking up the same transient data easily. Some of it can be explained away but some of it like the POSS-II is still unexplained.
Your AI said "The “objects” are single-appearance points on individual Palomar photographic plates — meaning they show up on one exposure but not on earlier or later ones.
- There’s no recorded motion or repeat detection.
- They don’t persist long enough to measure parallax, spectrum, or movement.
- The point-spread function (PSF) — how the spot looks — can look like a star, but so can dust specks, scratches, cosmic-ray hits, or tiny defects in the emulsion.
EDIT: “SDSS Stripe 82: This region was repeatedly imaged over ~10 years, allowing for multi-epoch analysis. Some objects that later vanished in Gaia or DES were detected multiple times in SDSS, suggesting they were not artifacts.” directly contradicting your AI answer.
“Pan-STARRS1 (PS1): Offers 12–14 epochs over 3.5 years. Some vanishing candidates were seen in several PS1 images before disappearing in Gaia or LSST pre-surveys.” again contradicting your AI answer.
Repeat detections are happening. Your AI is feeding you false information.
Last edited:
