The Skeptics Guide #1053 - Sep 13 2025

September 13, 2025 Unknown length

▶️ Listen to episode Download audio (MP3)

Quickie with Evan: Earth's Quasi-Moon; News Items: Guiding Tiny Robots, Tylenol and Autism, Music Choices as we Age, Wearable for Subtle Communication, Decreasing Science Scores; Who's That Noisy; Your Questions and E-mails: Gold Nanowire Gel Batteries, Otters; Science or Fiction

Jump to transcript

Listen and follow along

Speed:

Transcript

You're listening to the Skeptic's Guide to the Universe, your escape to reality.

Hello, and welcome to The Skeptics Guide to the Universe.

Today is Wednesday, September 10th, 2025, and this is your host, Stephen Novella.

Joining me this week are Bob Novella.

Hey, everybody.

Kara Santa Maria.

Howdy.

Jay Novella.

Hey, guys.

And Evan Bernstein.

Good evening, everyone.

So, did you guys hear the announcement today from NASA about about a possible biosignatures on Mars?

Uh-huh.

Yeah, I did, and I'm cautiously whatever.

Yeah, well,

I think it builds more evidence in the correct direction, saying that there was once life on Mars, yes?

It's a candidate, you know what I mean?

It's not confirmation.

So, they found basically they found speckles.

You know, like these are visible, like naked eye visible with the camera from Perseverance.

You know, they found rocks that have deposits which on Earth can be created with microbes.

Yeah, so yeah, there's two minerals that they found.

One's associated with decaying organic matter, and the other one is created by microbes on Earth.

But, of course, these minerals can be generated abiotically.

So, don't get really excited at all, really, about this.

But you would need high temperatures and acidic conditions to create these on Earth.

So, the most important but here is that there is no evidence currently for any high temperatures or acidic conditions in this area.

So what they've basically done is that they've reduced the non-biological origin probability a little bit, but you can't rule it out yet.

More studying to be done.

And I think chances are it's abiotic.

It was created either at low temperature, low temperature, which we're not sure

how possible that is in that area, but it's probably the

problem.

But here's probably

the bigger point, Bob, is we can't really know for sure unless we get them in an Earth lab.

Yeah, we we got to retrieve it.

We've got to retrieve those samples and bring them back to Earth before we could resolve that question.

Had they cut funding for that program to retrieve the vials?

No.

I think that's the first time.

Jay, did you talk about that on that prior news item?

Oh, yeah.

That was like scrapped, and then there was like talk about trying a different way to do it.

So, NASA's Mars sample return mission has faced significant funding cuts and potential

cancellation.

Oh, my gosh.

Cancelization.

Cancellization, yeah.

Did you hear the

NASA administrator, Sean Duffrey, give a little shout-out to Trump to try to

save, you know, to help NASA here?

He said, this finding by Perseverance launched under President Trump in his first term is the closest we've ever come to blah, blah, blah.

It's like he just threw that out there.

Like, wait, okay, I understand.

Well, what good is it going to do unless we get the samples back?

Yeah, well, I mean, at this point, I think any extra funding from the government would be, would have them, would make them overjoyed at NASA.

Absolutely.

And

saving any jobs, I think.

A little, like, you know, Schindler, like, I could have saved

10 more jobs if I just got a little bit more money out of Trump.

You know, that kind of stuff.

Or the legislative branch could do their job.

Right, and appropriately.

Well, that's silly.

What are you talking about, really?

But they did appropriate the money.

They should just assert their right to the purse strings, which is in the Constitution, rather than just laying down and letting

the executive branch steal their past much meaning anymore.

Well, back to the point of the significance of this particular news is that it says here: NASA's science mission chief Nikki Fox said, and this is a quote: it's the closest we've actually come to discovering ancient life on Mars.

Now, I understand we're inching that direction, but for her to say that seems pretty significant.

It's news.

It's certainly news, but I'm just not getting my hopes up.

I've been

burned too many times.

If there's an abiotic pathway, that's probably going to be it.

At least so far, every time that's turned out to be the case.

Right.

Just get back to me when you've got something that's three incremental steps and not just one.

But I mean, the bottom line is we got to get those rocks back from Mars.

And perhaps today's announcement is,

in a political sense, a way of re-establishing

the political will to get the funds to get this back to Mars.

It could be a strategic announcement.

Like, we really do have to get these samples back to Earth.

Bob, most importantly, I just want you to be happy, you know?

Oh, Bob's happy.

Evan, tell us about Earth's new quasi-moon.

Yes, I will do that.

May I start by giving a hat tip to Bob for allowing me to borrow the title of his reoccurring segment on the SGU?

Yeah, man.

I meant to actually read this, and I just got too hip-deep into my stuff.

So, please, I'm curious to see what this is all about.

Well, Bob, I know you can't see it right now, but I am raising my hat off my head in your direction.

This is your Quickie with Evan, and I'm going to let you know about Earth's newest moon-ish thing that has been added to Earth's slate of naturally occurring companions.

Earth, okay, so to review, Earth has one official permanent, solid, natural satellite object.

All those words.

That is what?

The moon.

And then we have these mini-moons.

You know what these are?

I think we've talked about these.

That's no moon.

Yeah, I think we've briefly mentioned them.

Yep.

They're like

temporary moons in a sense, right?

They're just kind of, oh, yeah, we're kind of sharing in orbit for a little while and then they move off, right?

Small asteroids temporarily captured by Earth's gravity entering the orbit.

and it could be just for a couple hours and some last several months but eventually the sun's pull reclaims it like mine mine mine exactly like the ring of power getting cast back into the fires of mount doom almost like that as a side note i looked there's this thing called uh kordalowski clouds k-o-r-d-y-l-e-w-s-k-i clouds clouds stable clouds of dust located at the l4 and l5 lagrange points okay yeah that's a good place to hang out yeah and they call them ghost moons ed Warren would be so proud of that.

Oh, God.

But they're just clouds.

They're not like real atoms.

They won't form.

They will never form.

They will never congeal together to form.

But they are clouds that orbit.

So that's what they're doing.

They're ghostly.

I love it.

They call them ghost moons, yes.

Back to mini-moons, though, for a second, and I'm going to get to the quasi-moon.

Earth doesn't, no fixed number of mini-moons.

Average of 6.5 of these, 6 or 7, orbiting Earth at any given time.

Now, here's the question.

Have we talked specifically on the SGU about quasi-moons?

I don't remember.

We did?

Yeah, we could.

Yeah, last time they discovered a new, one of these temporary satellites, we talked about it.

And that's what we've got again.

A brand new, well, a announced, newly, recently confirmed and newly announced

quasi-moon.

2025 PN7 is its designation.

This is courtesy of Earthsky.org, by the way.

Meet Earth's new quasi-moon.

Detected on August 29, 2025,

by the Pan-STARS Observatory

on Haleakala, Hawaii.

I hope I'm not sure.

Haleakala.

They looked at some data on this thing.

They had it back as far as 2014, from what I could find, and that they've added it all up and they've observed it recently.

But 2025 PN7, it has been co-orbital with Earth for 60 years now.

What?

I know.

It's been hanging around for a long time, but

only now is essentially effectively official.

And it's expected to remain for another 60 years before it will transition back to its original horseshoe orbit.

And I don't know that we've talked about horseshoe orbits, but that is a co-orbital motion in which a smaller body moves in a path that appears horseshoe-shaped relative to a larger body.

And they described it like two planets or a planet and an asteroid sharing an orbit around a star.

The smaller body's orbit oscillates, but its overall path, when viewed from the reference frame that rotates with the primary body, it forms a horseshoe.

So that's why it's called the horseshoe orbit.

But for a while, this thing is the Earth has it for now, and it will continue to have it for about 60 more years.

It's big questions.

Yeah.

How big and how far away?

Yes, those are my next two.

19 meters in diameter.

Far away, closest, 2.8 million miles, farthest, 37.2 miles.

So it's out there.

I mean, what?

That's like a a third of the way to the sun, practically.

37.2 million.

Yeah, right.

So 2.8 million miles.

It will not, yeah, so it's nothing that's going to come that close to Earth.

regular telescopes can't see it.

You have to detect it using some very special equipment.

I was wondering if it was worth even.

Imagine if we can just grab it.

I know.

Wouldn't that be?

I know.

That would be cool.

But we have it for.

Yeah.

But at least, you know,

we do have it in a sense for a little while, 60 more years in that far, far out orbit.

But

there were seven known quasi-satellites of Earth known before this, and now PN7 makes number eight.

So a heartfelt ahoy to Tony.

Ah, welcome.

PN7.

Welcome to the crew, man.

Welcome, and we're sorry we've ignored your presence these past 60 years.

It was nothing personal.

This has been your quickie with Evan.

I now send it back to Bob, who will send it back to Steve.

Back to you, Steve.

Thank you, Bob and Evan.

Welcome.

Jay,

I know people are developing

these tiny robots, but how can we get them to move where we want them to go?

All right.

Well, I mean,

this isn't like a, oh my God, I can't believe it research, but it is an incremental and very important piece of research for the idea that we would like to control very, very small machines to do a lot of different things.

I mean, just alone, you know, with the ability to help people, you know, like an example of it would be that it could deliver medication to a very targeted area in your body.

You know, instead of taking a pill that essentially gives your entire body the medication, right?

Yeah.

It could be like a targeted delivery that can go exactly where a tumor is, for example, or something like that.

So these

tiny robots, like historically, they

pretty much have to be dumb, right?

They don't have the space because they're microscopic.

you know they don't have the space for like a processor or a camera um onboard sensors and that type of stuff like that those components right now like they're they just can't be shrunk down that small so that means that these you know micro bots they can only respond to um you know like environmental cues if you want to make them do something useful like you want them to deliver drugs like I said, or clean a contaminated water supply, you're going to need to control them from the outside.

And you'd like to control them real time to figure out a way to make them navigate, to get to where they're going, where they need to go without them themselves having to do real processing.

That's it at its core.

So, researchers at the University of Pennsylvania came up with a different idea about this.

So, in a new study, William Reinhart and Mark Miskin, they describe a method they're calling artificial space-times, which sounds pretty interesting, and I'll tell you why they picked the word space-time.

So, what happens is they let their microbots follow a complex path without any central controller or onboard intelligence, like a real processor on board these tiny microscopic robots.

The trick is that they're actually changing their environment in just the right way so they will automatically go to where they want them to go.

And I'll explain how it works.

So, the microbots are made out of silicon, and they're powered by these tiny, tiny, tiny photovoltaic panels.

And those photovoltaic panels act like motors, but only under specific conditions.

So when light hits one of those panels, it produces an electrokinetic reaction, which actually pushes the robot forward.

So each robot has two of these.

So think of them as like a really small tank, right?

With two with two treads, one on the left and one on the right.

If both of the motors get the same amount of light, the robot will move in a straight line.

But if one side is brighter than the other, the motor on the left, say, which is getting more light, will move faster.

And

it'll make the robot turn.

And I know that sounds simple, but it was a very difficult thing for them to achieve.

And what they had to come up with is, you know, what they do is they project patterns of light onto the bottom of the Petri dish using a digital projector.

And from above,

it looks like a map, say, with light and dark regions.

And that the light and dark regions are very specific because as the little robots move through this, they can make them go to any point that they want to just by changing the light and dark areas of the projection.

Like I said, because they're reacting to the light that they're seeing.

So the robots can't see the map.

They're not aware of the map in any way.

All they're doing is reacting to their environment, which happens to be an environment that has light as a cue, right?

The light actually is making the whole thing happen.

So the bright regions pull them forward, the darker ones slow them down, and this allows them to change direction.

I watched the video of it, and it works remarkably well.

I know that they're incredibly small, but they look like they're moving pretty fast.

It's a pretty interesting thing, and it does kind of look like a flat tank, it's really cool.

The researchers were saying, like, you can think of it like the grooves on a record, like the robots are the needle, and they're blindly tracing the path that are etched into the landscape, in a sense, right?

Because that's what the light and dark regions are actually doing.

So, what makes this study different from the past ones is essentially the math behind it.

So, Reinhardt and Miskin noticed that the motion of the robots follows the same rules of light moving through

a curved region of space.

How about that?

So, the starlight, for example, that is bending, that goes too close to a black hole.

And what they did was that they made a connection between their robots in a light and dark space and essentially the way that

relativity works.

So, this connection that they made means that they can borrow mathematical tools from general relativity relativity and optics to describe or design their light fields.

And it tracks, it works.

Interesting, yeah.

So they're using equations now that the big man came up with.

You know what I mean?

They don't have to guess, they don't have to do a lot of testing.

They can just calculate exactly how a given pattern will guide the robots.

And it could be a very complicated environment, like mazes, branching channels, whatever it is.

And they can come up with the light pattern as a mathematical equation.

They make it, and it works, and they're doing it.

So in one experiment, experiment, they designed a light field.

It caused the robots to patrol a circular path, like security guards walking around, right?

In another example, they built a maze and they projected a pattern that let multiple robots find their way to a specific area that they chose without ever touching any of the walls of the dish.

And I think the most impressive one was when they created a sorting field, a single static light pattern that made

different robots exit a junction through different different paths purely based on where they started.

So it's complicated and it's working.

And the important thing here is that they didn't have to make any real-time adjustments, meaning, you know, let's say they're looking at the activity of the robots.

The robots are just doing what they're going to do by

the way that they're designed, meaning they have the photovoltaic sensors that produce the charge.

And it really comes down to the design of the light and dark areas.

They don't have to tell, hey, robot number 25.7,

turn a little bit to your left.

There is none of that real-time communication necessary.

The robots will go to where the light directs them to go.

So once the projector turns on, that pattern stays fixed and the robots don't need any individual instructions or monitoring.

It's really cool.

It makes it inherently scalable, and you could release thousands of microbots and they'll all follow the same rules all at once.

So now

you're controlling thousands of or even more of these tiny robots instead of only being able to communicate to a handful of them.

And they can't even be as small as they are if they're being communicated to.

I feel obligated to tell you, this is just a lab setup.

The robots aren't, you know, they're operating in a shallow

petri dish with liquid

of a certain height under carefully

manipulated lighting.

In real-world applications, there'd be a lot of noise in the system, right?

There'd be random disturbances.

There'd be imperfect

manufacturing.

the projected light field today, like it can't work deep inside of a human body or in a murky river, right?

So there's restrictions, of course.

But again, this is just the beginning, and it's a really cool thing that they came up with.

Going past the idea that it's a leap, you know, this is a conceptual leak that they came up with.

What would the real-world applications be for things like this?

Like, why are we, why are they even doing it?

The scalability aspect here is really important because the current systems that exist can only control a few of these microbots at a time because each one of them needs constant tracking and external commands.

This method uses a single static light field and it can handle swarms of these things.

So it's no problem.

It just makes it super easy that way.

The simplicity brings down the cost dramatically.

So the robots are cheap to make.

They're simple.

They don't need onboard computers.

They don't need onboard power sources.

It's just all there is no intelligence.

It's all by the design of the light field.

I don't think we should just shrug off a lot of these interesting little things that happen like this because it took thousands and thousands of these to make an iPhone, right?

It had to start somewhere a long time ago with people coming up with taking small steps in an unknown direction that later got picked up by other engineers and using research that was done 50 years ago to help them figure out how to do something today.

So I think this one seems pretty obvious that there's a cool benefit to it with other, you know, other labs taking this in other directions as well.

I just think, overall, like this type of research is the exact type of things that need to be done.

And this is why we shouldn't cut science funding because we want people to do wacky stuff like this.

Because, again, it's all part of the aggregate knowledge base that other

researchers and scientists and people in the lab can use to make even bigger and more cool things.

Eventually, maybe they will be able to

put these things in somebody's body and it goes right to where they want it to to do something very specific.

Yeah, I mean, this

kind of technology is one of those, I think, for now, it's like a sleeper technology.

They're working out the basics, doesn't really have a lot of applications.

But if we get to this threshold, you know, in terms of like the level of control and the level of functionality, there could be some amazing applications of it medically.

Yeah, it's important.

But we just don't know at this point.

I think it's really important for people to understand the idea that nothing, no new technology today is just invented out of whole cloth and they have a finished, exquisite technology.

It is dependent on just millions of hours of human research that came before it just to find out, you know, to figure out small things.

You know, make you know all these little things.

Okay, look, we know how this functions now.

We know what the chemical reaction does over here.

You know, not one of them by itself really means that much, but it's the aggregate knowledge base that allows more complicated things to develop later on.

So we need researchers doing stuff like this.

Hey, Jay.

Yeah.

Did you see the most recent last week tonight?

Because that was like the thesis statement of John Oliver's main piece.

No, tell me what he said.

I mean, it was basically just that.

We can talk about single studies all we want, and we can sort of poke fun at studies that ostensibly look like they're not doing something important for humanity.

But ultimately, all of the great discoveries we have are made up of thousands of human hours and hundreds of published studies.

And it all has to work together.

Aaron Ross Powell, Jr.: And it's often researchers pursuing some quirky interest or curiosity with no idea of what it was going to lead to.

Definitely.

I mean, look at it like a web, right?

Like you could have all of these different epicenters on that web that are discrete pieces of research and stuff.

But eventually, they do start to connect and people will start to go, oh, we could take from here, take from here, take from here.

And exactly what you said his show was about.

Like, absolutely.

Yes, it costs a lot of money and it takes a lot of time.

But that's where science gets all its cool stuff from.

Right.

So we have to be okay with that.

And that's, I think, the sad part here is that the government is just stripping all of this research, anything like this.

It's just so much of it now is being destroyed.

And we're not finding it.

Yeah, we're destroying our scientific research infrastructure.

It's terrible.

It sucks.

But anyway, this is cool.

Don't get depressed.

Be happy.

This is cool.

Little robots now will drive around on a light beam, Steve.

Thank you, Jay.

So, do you guys remember back in April when RFK Jr., sorry to bring me down with RFK Jr.

when he predicted that, and this is in quotes, by September, we will know what has caused the autism epidemic and we'll be able to eliminate those exposures.

Let me check the calendar.

It's September.

It's September,

and he hasn't been able to surpass decades of research in five months, shockingly.

Of course, you know, we all suspected that he's making statements like that because he already thinks he knows what, quote-unquote, those exposures are, right?

You know, you can't start from scratch and complete any kind of reasonable scientific study in five months.

Like, we're not going to unravel whatever aspects of autism remain to be unknown in those five months.

He was was just setting up this idea that he's going to be blaming vaccines for autism.

So here we are, it's September, and there's a report from the Wall Street Journal and some other outlets that an upcoming HHS report

by RFK is going to tout a link between Tylenol, acetaminophen, or paracetamol in Europe, and autism risk.

Exposure to Tylenol in pregnancy, in mothers, and the risk of their child having autism.

Isn't that debunked already?

It's already debunked, which is just the great, it's great.

The leak of that that's what the announcement's going to be gives us, you know, science communicators the opportunity to pre-debunk it before it even comes out.

And maybe they won't bring the paper out, who knows?

But let's just review the evidence really quickly, just so you have the information in case family members like, oh, I heard there's a link, blah, blah, blah.

But let me back up a little bit because I always like to draw generic critical thinking, scientific literacy lessons from these individual stories.

It's important to recognize that whenever you look for correlations, you'll find them.

Pretty much, you know what I mean?

There's going to be correlations pretty much everywhere.

It's like if you're looking for patterns, you will find them.

And oftentimes, when people are looking, like they're doing observational research, meaning they're not doing a controlled study where they're they're randomizing subjects and either giving them a placebo versus a treatment.

They're just looking at the world and saying, oh, if we look at this cohort of people, what happened to them?

Like, what did they do and what happened?

They're just observing what's happening out in the world.

There's lots of confounding factors, obviously, because you're not controlling for everything.

There's lots of variables.

And if you just, you know, slice and dice the data in a few different ways, chances are pretty good you're going to find some correlations.

And that's often how preliminary studies are done.

You may look for 20 correlations, or you may look for a hundred correlations.

And this is not really to confirm any kind of cause and effect relationship.

You're literally, you might just be, you're just on a fishing expedition.

You're just saying, hey, is there any signal out there?

And then that is meant really only to create a hypothesis that there may be a correlation there.

Then you have to confirm it with a fresh set of data where you're looking just at that, right?

Also, these kinds of studies tend to have a massive false positive.

That's just by their design.

Again, this is like screening tests are meant to have false positives.

That's how they are designed, statistically, et cetera.

Because you're screening, right?

You want to capture everything, and then you follow up with a more careful or detailed or precise analysis to see if what you're finding is real.

Does that make sense?

So, what that means from a practical point of view is that there's always going to be these preliminary evidence for all kinds of correlations.

Of course.

And not only that, it goes both ways.

If you just look at, oh, let's look at some exposure to some food.

If anything gets worse, it's a risk.

If anything gets better, it's a cure.

You know what I mean?

It's like anything can happen, and it becomes either, oh, this will prevent this disease, or this will cause that disease.

It's meaningless.

It's noise in the background.

That's meant for researchers to say, okay, let me pick up the ball and see if this is real, right?

To really confirm or deny this hypothesis.

That's the important point there, too.

You have to actually be doing hypothesis testing.

You can't just go, let's look at big data and see what trends in a certain direction because everything's correlated with everything.

Yeah, that does the data.

Yeah, right.

You're just, you're going to find things.

That's not a confirmation.

You have to then later confirm it.

Okay.

Although there are analyses you could do to see how strong that correlation is and how and if it's implying a particular or favoring a particular causal relationship.

There's always a lot of details in there.

Yeah.

All right.

But if you're if you're RFK Jr., right, or you're a wellness warrior or now, I guess, a Maha loser or whatever, right?

If you're somebody

whose career is based upon this kind of thing,

eat this superfood to be healthy or avoid this cancer risk or whatever, then all of these preliminary correlations are gold, right?

I mean, you could dumpster dive through all of these preliminary, mostly incorrect correlations to weave whatever tail will sell your supplement or make people afraid or whatever it is that you're trying to do, right?

Build your guru career, you know, as a wellness idiot, right?

This is the world in which RFK Jr.

lives.

He clearly has demonstrated he has no idea how to properly interpret the scientific literature.

And he just cherry-picks these scary preliminary data to tell whatever story he wants to tell.

And of course, sometimes he will cherry-pick studies that were done by pseudoscientists that there are terrible studies just to make whatever point he wants to make, like that vaccines cause everything, for example.

Okay, so yes, there is, having said all that, there is preliminary data where if you just look at the raw data, there is this tiny effect really, there's this tiny correlation between mothers using Tylenol in pregnancy and increased association with either autism or ADHD or other neurodevelopmental disorders, right?

However, so that's almost certainly what RFK Jr.

is going to be touting if he does go through with this report or what others have stated that there's this correlation.

But the thing is, we've already done the follow-up research.

In this case, right, if we hadn't, I would be saying we have to wait, we have to do the follow-up studies, we don't really know, this is preliminary.

But now I get to say we've done the follow-up studies, right?

So in 2024, which is fairly recent, there was a pretty high-quality review.

They looked at

the Swedish national database, and this review included data on 185,000 children.

So it's pretty statistically robust.

Again, this is an observational correlational study, not a cause and effect study.

So they did find, yes, there's this very small increased risk of neurodevelopmental disorders with Tylenol use in pregnancy.

But there are two reasons to think that this is not a real cause and effect.

One is there was no dose response curve.

right so if a drug is causing a benefit or a substance is causing a risk or harm, you would expect that more of it would cause more of a harm, right?

We call that a dose-response relationship or a dose-response curve.

And so, it's good to include that analysis.

So, what they did was they said, okay, if we go, if we compare the lowest 25th percentile of exposure to acetaminophen, and then the middle, medium, you know, 25 to 75th, and then the highest 75th percentile, greater than 75th percentile, is there there an increase?

And there wasn't.

It was a flat line.

It was random.

There was no dose response curve.

So that really argues pretty strongly against there being any kind of causal effect here.

But even more definitive than that, the authors did what's called a sibling control analysis.

And this is a way to control for confounding factors.

Remember with the observational data, the biggest weakness is that there are confounding factors, meaning there are factors you are not controlling for and you are not randomizing and that could be seen in the data.

And the simplest way to think about this is if A correlates with B, you can't assume, if it's an observational study, that A causes B.

It's possible that B causes A.

It's possible that C causes A and B.

So it's possible that pregnant women who are more likely to take Tylenol

have something else going on that could also make them more likely to have a child with a neurodevelopmental disorder.

So one way to control for this is to do a sibling control analysis.

What you do there is you use the subject's sibling as a control.

And so you have like one sibling with

the exposure of interest and one sibling without the exposure of interest, right?

So you could say during my mother's pregnancy with me, she took Tylenol, but with my mother's pregnancy with Jay, she did not take Tylenol, right?

So you've kind of isolated the Tylenol exposure from all of the familial and genetic and some of the environmental factors.

There's enough of those kinds of case

candidates to make a study on?

Absolutely.

And so they did that kind of sibling control analysis, and the correlation completely goes away.

There is no correlation once you control for that basic familial confounding factors, right?

And that's for any of the three, any neurodevelopment, either autism, ADHD, or other

neurodevelopmental disorders.

So that's pretty definitive.

That means that

there is no real correlation between Tylenol use and autism or other disorders.

And the authors conclude that prior studies showing a possible link may have been attributable to familial confounding, right?

To these confounding factors.

So this is pretty robust evidence.

It's a large study.

It's well done.

And again, those two things, lack of a dose response curve, the effect goes away when you control for confounding factors.

That's always the death knell in an observational study.

Like when you control for confounding factors, the effect goes away.

It wasn't real to begin with, right?

It was due to the confounding factors that you're now controlling for.

So, but this is like three layers too deep for somebody like RFK, right?

Who doesn't care about this?

He's just saying, oh, look, there's a possible link that's enough to scaremonger about it.

Look, I kept my promise.

I found the exposure that's causing our autism epidemic, or whatever.

He'll do a victory lap, is what I predict.

So we'll see what actually happens.

But it's good that the data is already there, that

science communicators are already getting out there saying, nope, there's no evidence for this.

The evidence does not support a link.

The other issue here is that Tylenol is really the only safe way to treat fever in pregnancy.

Scaring women against using that could have harm.

It's not just like, oh, who cares?

So generally speaking, we tell pregnant women not to take any medications they don't absolutely have to take.

That's sort of standard procedure.

But if you're having a lot of pain, that's not benign.

That is a potential risk for a pregnant woman.

And if you may not be getting enough nutrition or whatever, there's all kinds of things that you may be spending too much time in bed or whatever.

That's a risk factor.

Also, fever, you know, can be a risk factor.

So telling women don't treat your fever or pain during pregnancy is not a benign thing to do.

No, it also just perpetuates, I think, these, I don't want to even say decades, centuries-long

biases in healthcare against treating women's pain.

Just being like, just deal with it, you know?

And even if,

let's say, a woman who is pregnant is having a severe pain that is not linked to anything dangerous, right?

Even if she doesn't have a fever or it's not indicative of some sort of harm, it will cause bad psychological outcomes.

She will, like it will contribute probably

to depression during the pregnancy.

It will contribute to negative emotions about her pregnancy if she feels unwell and she doesn't have options in front of her.

It's like,

I don't know.

It just feels like another one of these ways to blame mothers like we used to do in the 50s.

Totally.

People have brought that up as well.

This is just an extension of the idea that mothers are to blame for their child's conditions.

Even when it's, we know from mountains of evidence that autism, for example, is dominantly genetic.

That doesn't mean to say there are no environmental factors that might affect that, but it's dominantly genetic.

It's not dominantly

environmental, which RFK seems to believe, again, against all evidence.

And also keep in mind, like, Tylenol is our first-line treatment for migraines.

We're not just talking about, oh, I have a, I stugged my toe or something.

And, you know, a migraine during pregnancy could mean significant disability, nausea and vomiting, things that are, again, not benign for the pregnancy or the mother themselves.

And so everything is a risk versus benefit, right?

Everything in medicine is risk versus benefit.

So it's not as simple as saying, well, just avoid all risk, you know, at all costs.

It's like, life is not that easy.

You avoid one risk, you incur another, right?

If you're by avoiding that risk, you're also avoiding the potential benefit.

And so you have to make a more rational sort of decision about where is the optimal balance, you know, of risk versus benefit

for each individual patient in each individual situation.

This kind of fear-mongering, based upon misreading of the evidence, is harmful.

It is literally harmful.

And I don't even, I mean, like, I feel like it's even charitable to call it misreading of the evidence.

I think it's confirmation bias in a group of people who fundamentally are pushing an agenda that's anti-medicine.

Completely.

Yeah, it's basically this weird sort of paleo viewpoint that somehow more medicine is worse.

And if you can live a, quote, clean lifestyle.

Yeah, it's the clean living bullshit.

Yeah, avoid any pharmaceuticals and avoid any sort of intervention, that it's going to be healthier for you.

I mean, it's the same, it's the anti-vax, it's all of it.

And it's just so dangerous.

Absolutely.

All right.

Well, Kara,

you're going to talk next about the music choices we make as we age.

Yeah, I'd love to take a mini poll just of the rogues today.

How many of you...

Okay, first and foremost, do you tend to listen to more music from when you were younger or more music from the recent past?

Younger.

Younger, yeah.

I'm influenced by my daughter in this regard.

So I don't know how typical I am, but one of the things things Rachel and I do, we go to concerts together.

So a lot of my music is more contemporary.

You're listening to what she listened to.

Sort of by default.

Exactly.

Right.

And I do like keeping up with what she is listening to.

I agree.

It's our way of bonding.

The second category of music I listen to is music my daughters listen to.

Yeah.

100%.

And would you say that that makes up like half of your listening?

Yeah.

Most of your listening?

Yeah.

I would say probably half and half my youth, their youth,

very little

current music that

I'm not being exposed to to somebody anyway.

I'm definitely a blend.

I'm a 50-50 blend.

Okay.

And then for, let's say, for like Jay and Bob, do you use any sort of service like Spotify or something like that that recommends music to you?

I don't let anybody recommend music to me.

So you only listen to what you want to listen to.

That's right.

Yeah, I mean, you know, I will hear stuff that my kids like.

And, you know, there's a few recent artists that I definitely definitely like.

And I'll, every once in a while, I'll hear some of their new stuff.

But I mean, in my mind, you know, I just can't find anything that's better than the stuff from the late 60s, 70s, 80s, and 90s.

Like, I, you know, I just, I don't know.

Is it my age?

I don't know, Kara.

And so I'm going to talk about.

a study that is not going to tell us the why.

It's going to tell us more the what.

And it's going to tell us something that we already knew.

But this is the first time, apparently, that we have like scientific evidence to back up the claim that as we get older, we focus on fewer and fewer favorite songs and our musical taste narrows.

But what I love is that the authors, instead of talking about our musical tastes narrowing, they use I don't want to call them euphemisms, but I guess their language bias is probably because they're older as well.

Our musical taste becomes more refined as we.

And so here's what they did.

It's actually a really interesting approach.

So it's three.

More seasoned.

Yeah, more seasoned.

And I want to hear what your answer is too, Kara.

Oh, I'm 100% the same.

Like, I listen to stuff I listen to in high school all the time.

All the time.

And some kind of vintage stuff from before that.

I do listen to some current and modern music, but by and large, I would say that, so I work at a gymnastics gym and I'm definitely the oldest person.

Well, I think I'm technically the second oldest person there.

And oftentimes, when they're playing music, I'm like, I've never heard this before.

I have no idea who this is.

They'll mention the name of the artist, and I'm like, okay.

Sure.

Yeah.

And I'm, by the way, 41 years old.

Like, I am, I'm not old.

I'm not young.

Right in the middle.

But so these universities, this international study, what they decided to do is try to understand people's music habits a little bit more, especially focusing on like different artists and different genres that individuals listen to.

And the way they did it is by looking at big data from a service called Last FM.

This is a music service where you can sort of feed in your listening habits, like from a place like Spotify, and build out a profile so that you can get more sort of more of an analysis of your own musical habits and tastes.

And when people make profiles on Last FM, they put in their age.

Now, we don't have a lot of data on how these algorithms work over long periods of time because they haven't been around for that long.

So, this is one of the earliest studies that was able to look at 15 years

of data, and they looked at over 40,000 users.

So, we're talking like almost 550 million plays,

and they looked at over a million different songs across those 15 years.

So, they wanted to see how people's music listening habits changed over time.

And they were especially curious because this was actually published as an extended abstract in a proceeding, like a conference proceeding, at UMAP, which is is user modeling, adaptation, and personalization.

So these are, you know, scientists who are interested in how these different services feed you what you want to hear.

Like, how do these algorithms work in such a way that the end user is enjoying their experience?

And basically, they found that programs like Spotify or

companies like Spotify, they're not targeting older people in a different way than they target younger people, but older people's habits are different than

younger people's habits.

Younger people are really interested in

new music.

They're really interested in novelty.

They're interested in trying on different hats, different personalities, different genres.

They want to hear stuff they haven't heard before.

Older people, by and large, not everybody, but by and large, they tend to, and this is their quote, which I think is lovely,

middle-aged listeners appreciate a balance between new and familiar, while older listeners want more tailored recommendations that reflect their personal tastes and nostalgic reminiscences.

So,

what they found is that older people's music habits, by and large, focus on songs from their youth, that people do go back and listen to songs from when they were younger.

And as they put it, musical taste becomes more, quote, unique the older the listener is.

So, like, they claim that teenagers will often find a lot of common favorite songs with other people their age even if they're sort of in different cliques or you know go to different schools whereas as you get older your musical tastes may become more specific and it may be harder for you to meet other people your age who are into the same kind of music you are You know, like I have an ex-boyfriend who just listened to like death metal all day, every day.

I mean, he would hate hate that that's how I'm classifying his musical taste because in his view, it was incredibly varied, right?

But it was all just hard.

He listened to a ton of hardcore music.

And, you know, he was in his mid-40s.

I don't think he had a lot of friends who were listening to music.

that was as intense as the music that he listened to.

He had a handful of them, but for the most part, people were like, bro, I can't get into this.

It's like loud.

Can you turn it down?

And I see that across the board.

You know, I've I've got a lot of friends who still listen to punk rock.

They love their punk rock.

And for my friends who aren't into punk rock, they just don't like it.

And so it's hard to find commonality and to talk about music with your friends and to go do musical things with your friends unless you maybe went to school with them or you already bonded over those musical similarities.

But definitely kind of making new friends as an adult, you do learn that you might have a ton of stuff in common.

And then you're kind of surprised by how different their musical taste is than yours.

I found that that's happened a lot.

Kara, I just reconnected with someone I hadn't spoken to since 1991.

I mean, this year.

And this was during, this is my last year of college.

This person was into heavy metal.

I mean, real heavy stuff.

I asked him kind of what he's listening to now.

He's like, oh, Billy Joel.

Just like, what?

No way.

It threw me for a loop.

I had no idea.

That's interesting.

I wonder how much of it is just changing music itself versus our changing life situation.

Like, when we were younger, we were in school and we had lots of exposure to people who were like playing new albums and you would get to hear them.

And you know what I mean?

Like there seemed like there was a set of new stuff coming out that you got exposed to and you could sort of choose what you liked.

And now I wouldn't know where to go.

Like there's just so much out there.

Oh, it's and the stuff you hear on the radio is largely crap, you know.

Well, but that's also an older person's perspective.

You have to remember that, too, because, like, if we take any cross-section of time, the older set is going to say that that newfangled music is crap.

But it's not just that, because when I had, when my daughters were at that stage, when they were still living at home, they were in high school or whatever, and they were sampling a lot of different groups.

First of all, they were into a lot of older groups.

They were listening to foreigner and, you know, like stuff from our generation.

Then they were listening to new groups that I otherwise wouldn't have heard about.

So I was able to select the ones that I liked.

And then I got into, you know, Muse and

Arctic Monkeys and stuff like that that I otherwise never would have gotten into.

Which is now old again.

I know, it's now old again.

So there's got to be good bands out there that I would get into now if I had the ability to get exposed to them.

But I just don't think it's the pop stuff that you're hearing.

It's more like the next layer stuff that's maybe not alternative music.

It's alternative stuff.

Well, but even if we're talking about one genre like alternative, which has a million sub genres within it, I think probably what

not just what these researchers are kind of getting at, but also what individuals working behind the scenes at companies like Spotify and Pandora hopefully are focusing on is the idea that people get dialed in to their tastes and they like the nostalgia of their youth, but there is always new music that has some sort of connective tissue, some sort of flavor, some sort of vibe.

And it is hard to pick out what those variables are, right?

Is it a tempo?

Is it like a beat?

Is it

the timbre of somebody's voice?

Is it sort of the sounds of the guitars?

But there was a lot of things.

Why isn't Spotify doing that?

Why aren't they

stuffing?

I should like.

They are.

And so there are settings within all of these different apps where you can have a mix and then you can put it on like a smart mix or whatever they call it, where it will recommend things to you.

But what these individuals are saying is that they're doing that to an 18-year-old and to a 65-year-old the same way.

And what they probably should be doing is narrowing that algorithm for the older individuals so that there's not as broad a mix being kind of thrown at them because they're less likely to to go, oh, I like this.

And they're more likely to just downvote everything that doesn't sound just like what they liked before.

They're just not intentionally searching for new music.

And so it's not to say that older people wouldn't engage with newer music, but that newer music probably needs to be more narrowly focused on their established tastes as opposed to having, I guess, a really eclectic, let's throw stuff against the wall and see what sticks approach, which works with younger people.

What do you think, Che?

I mean, you know, if it weren't for Rick Beado, I would be agreeing with Kara.

I mean, it seems like every generation kind of has this experience.

The bottom line is, I think that there's very, very real world reasons why music, the quality of music in general is not as good as it used to be.

And this comes from a deep analysis of the complexity of the popular music today.

So I think it's kind of both.

I think that because of companies like Spotify and Pandora and that whole streaming model, it's denutted bands from being able to make money.

And it's changed the industry so much that, you know, there isn't the same quality of bands out there.

You know, the vetting process is not the same.

There used to be promoters and people that their job was to find new talent.

And

there was a cream rising to the top effect here.

Yeah, I disagree.

I think now we have a cream rising to the top effect.

It doesn't rise as high to the top, but we're not going to be able to do it.

But isn't it also buried in more stuff?

There's just so much out there.

But that's the thing.

To us, it's buried, but not to people.

You know, a 17, 18-year-old right now who is on TikTok knows who the hot people are, who is like using SoundCloud, who is looking at these different outlets, and they're finding people who have a completely democratized approach to music.

There aren't the barriers.

to entry that they used to have.

And yes, there is so much more music out there.

And it's harder for us like old people to figure out how to dig through.

We're just not plugged in.

Is that what you're saying?

We're just not plugged in.

But the thing is, I would argue, Jay, that actually removing all of those barriers to entry, yes, it might make it harder for you or for me to find that incredible music.

You know, pop's going to always be pop.

There's always going to be the stuff where we're like, I don't get this.

I don't get this top 40.

Like, why is this so popular?

But when you can, when anybody can self-produce an album and publish it, the quality of music is going to go up.

It just is, because there are going to be people who didn't have access before that now have access to the music.

I think that's largely the pop music that Rick Piedo is talking about.

And sure, I mean, there could be millions of diamonds in the rough throughout all these streaming services that I just don't have a pathway to get to.

Yeah, and I think that's really what this is about is how can these music services improve that pathway for older adults who just don't want to dedicate that much time or energy.

But I think we all have to be careful not to fall victim to, I mean, I guess this could be a name-that logical fallacy, where we tend to compare the best of our tastes to the worst of other people's tastes.

That's not fair

because I hear it a lot, especially with the like older people argument that, like, in my day, music was amazing because the Beatles.

Yeah, the Beatles were great, but now what is this crap?

And it's like, well, but you're picking like a top 40.

No, I agree with you.

I just want people to tell me who are the great bands of today so I can listen to them.

Yeah, why would it be so hard for them to, for people like us to even find out about them?

If they're that freaking good, you'd figure we'd know about them.

And when we're going to get emails with people telling us, giving us their recommendations and do it.

I want to hear them.

I would love to be plugged in to ways of finding great new music.

This is how you do it, right?

I love the stuff from the old days, but I'm getting bored with it.

Yeah, and here's the thing.

Here's the big difference, right?

Teenagers have more time than we do.

Yeah.

They put their time and effort into this.

Young people go to festivals.

I just don't go to festivals anymore.

No, I know.

When I was young, I would listen to the radio when I was in the car.

Now I listen to the news.

Exactly.

And like when I want to see a band, which is rare anymore,

and that's not because I don't love music, but I need to have my earplugs and I need to have a seat.

I don't want to be standing all night.

But when I want to see a band, I'm going to go see a band I love.

And I want to, you know, and I actually am annoyed if a band that I love just put out an album and I haven't really had a chance to get into it.

And that's all they play at the show.

Like, I want to hear the stuff that I love.

Whereas when I was young, I would go to huge musical festival, like music festivals that had like 15, 20 bands playing.

And I might go, oh, I've never heard this before.

This is pretty good.

Yeah.

I don't do, I don't seek that.

I don't seek novelty out the way I used to.

Yeah.

And so we've got to remember, too, it's a lot of it.

This is user error.

Yeah.

I totally agree with that.

Yeah.

You know, 41 is is kind of old.

It is.

Listen to

Sing Like Sinatra.

It's crap.

Get off my lawn.

And we didn't even talk about AI creating music.

Well, then the AI slop is going to be, we're going to get buried in that for sure.

Yeah.

Oh, my gosh.

There's going to be millions of songs.

Does anybody even

do people like AI slop?

Nobody actually likes AI slop, right?

No, but it's out there.

I know it's out there and it's burying everything.

But are people actually choosing to listen to it?

Or is it like that U2 album that was on the IBM?

They might listen to it and not know they're listening to it.

Yeah, that's

the favorite music of all the stuff.

And the thing is, it's going to get just good enough that it's like mediocre.

No, it already is.

It's in the range of the shit that's already out there.

You know what I mean?

I've definitely come across songs before where I'm like, I don't think this is a real band.

It's going to explode the low end of the music distribution.

Make it even harder to find the good stuff.

Well, everyone, we're going to take a quick break from our show to talk about our sponsor this week, Quince.

You know, here at SU, we love Quince.

They have all the fall staples that you will wear all the time, like super soft, 100% Mongolian cashmere sweaters, starting at just $60.

And they partner directly with Ethical Factories and Skip the Middlemen, so you get top-tier fabrics and craftsmanship at half the price of similar brands.

But it's not just clothes, they've got all sorts of things for your home and jewelry, too.

Right, Jay?

So last Mother's Day, I gave my wife those earrings.

Remember, I told you about the Quince earrings?

So my wife goes ahead and loses one of them.

So the bottom line is she calls up Quince and they said, send us back the one you have and we will send you a brand new pair.

Who does that today?

That's amazing.

Great customer service.

This brand is fantastic.

That cinched it for me.

I love their clothes, but their customer service is fantastic.

So keep it classic and cool this fall with long-lasting staples from Quince.

Go to quince.com slash SGU for free shipping on your order and 365 day returns.

That's quince.com slash sgu.

Free shipping and 365 day returns.

Quince.com slash sgu.

All right, guys, let's get back to the show.

All right, Bob, tell us about mind-reading wearables.

I'll premise this by saying the article you late sent me, I did not like.

Yeah, actually, there was, There's not a lot about this out there, so I'll take some of this with a grain of salt.

Some of the text seemed interesting, but let's just jump in and we can discuss it.

Alter ego is a startup spin-off from MIT, was recently in the news again, describing a wearable device that people that allows people to silently communicate back and forth with either an AI or a computer or other people just by thinking the words.

They claim their primary focus is to help people with speech disorders, but it also has a plethora of other potentially interesting uses as well.

The MIT website describes the device like this: Alter Ego is a non-invasive, wearable, peripheral neural interface that allows humans to converse in natural language with machines, artificial intelligence assistance, services, and other people without any voice.

So, yeah, so this started as a project at the MIT Media Lab in 2018 and was spun off officially, I guess, this year, as a for-profit company called Alter Ego.

Arnav Arnav Kapoor is an MRT graduate, Harvard Medical School researcher, and of course CEO of Alter Ego.

So he said this.

He describes his technology at a TED Talk back in 2019.

So it was what, six years ago.

He said, imagine talking to yourself without vocalizing, without moving your mouth, without moving your jaw, but simply by articulating those words internally,

thereby very subtly engaging your internal speech system, such as your tongue and the back of your palate.

When that happens, the brain sends extremely weak signals to these internal speech systems.

Kapoor then describes next that they had developed a sensor that they developed for your face and neck that could pick up those subtle signals, allowing an AI to turn these patterns into the words that they represented that you were consciously sub-vocalizing.

The AI then silently answers your query with

whatever it may be by using bone conduction that travels to the inner ear that you can then hear seemingly normally, but without apparently drowning out, apparently, without drowning out whatever else that you may be hearing at the time, right?

So you could, it's not going to disrupt what you may be hearing necessarily.

So that's how alter ego works in a nutshell.

That process kind of makes sense how this works.

One question I had, I had here, you may have as well.

Why does this inner speech even send these faint muscle signals in the first place?

That they don't even move the muscles, but these signals are being sent when you're just thinking about words, specifically thinking about why does it mean...

That's the motor planning area, Bob.

Yeah, well, yeah, there was a lot of technical reasons why that's the case.

One way to describe it that made some sense, and see if you agree with this, Steve.

One answer is that it's ultimately faster and cheaper for the brain to allow some of this signal leakage to the speech muscles than it would be to perfectly isolate those muscles from inner speech.

So we're just like, oh, yeah, some of the signal gets out.

It doesn't do anything.

It doesn't really, it doesn't move any muscles really, but it's just like, it's easier to do, to do that than to really isolate

the muscles.

I would go further than that because

I wrote recently on Neurological Blog about a study where they're charting the brain process.

This is in a mouse brain, but they basically totally mapped out everything that's happening in a mouse brain when the mouse is doing certain tasks, right?

Some specific specific decision-making tasks.

And what they found was that the sensory and motor areas were far more involved in that process of thinking and decision-making than was previously suspected.

So we actually think with our motor cortex, if that makes sense, to some extent.

It's involved.

That's why we're so embodied, you know, and

it's

the circuits are all related.

So it makes perfect sense.

Like, if you're thinking about yourself hitting a tennis ball, the muscle, the motor part of your brain that is involved with that task is lighting up even if you're not moving, right?

Right.

it's actually a nice feature that we can exploit to do this sort of thing, right?

Exactly.

And that's exactly what we're doing.

And Steve, it also reminds me of, I remember studies that I read many years ago.

Hopefully it's probably still correct.

That when you're dreaming and doing things in your dream and moving, that they could detect

those same patterns in your brain that are being activated because

you are moving in the dream, so your brain kind of thinks you are moving in

a real way.

When you are thinking of a picture,

the same part of your brain lights up as when you are looking at the picture.

It's the same thing, right?

So it works with perception as well.

So Kapoor says at this point in his TED Talk, he was very clear that he said, just to be very clear, the device does not record or read your thoughts.

The control resides with the user.

And so, yeah, he specifically said that it seemed like he was making a very important point.

And of course it is.

You don't want to think about some device actually reading your thoughts and you having no control.

Because that's obviously quite scary.

And that's not the case with this specific tech.

And this idea, though,

ties into the big distinction between this specific technology and most brain interface technologies that we've discussed in the past.

Directly recording brain activity, you need something that's somewhat invasive.

Either it's going to be a full-on implant in like under your skull, in your brain somewhere, maybe even on the surface of your brain, but in your beyond your skull.

But also, it could be all the way down to something like electrodes that are attached to your bare scalp, right?

is that accurate enough, Steve?

That's kind of like the range of invasiveness that you'd be dealing with with the

conventional electrodes.

There are stent electrodes in your, in the vessels in your brain, then there's like brain surface electrodes and then deep brain electrodes.

So, those are obviously quite problematic.

They're effective for that type of technology, but it's invasive.

It's invasive, and they really need to nail down how to make it less invasive and more permanent because

even connections in the brain

eventually will drift and

need to be addressed.

So, alter ego is less invasive than in two ways.

It slips on easily and it gives users more control over what is shared.

It's only things that you are specifically and volitionally saying or thinking with your mind.

So it's less invasive in those ways.

And this technology, I guess it's not a surprise.

It fits very well with what Kapoor has, his earlier beliefs about the relationship of humanity with technology.

He's long advocated for technologies and AI to be developed that integrate and extends human capabilities, but to augment us instead of diminishing us, as he says, or replacing us.

And I'm sure many people would agree with that these days, with what's going on with AI, especially with artwork and things like that.

So Kapoor describes alter ego as giving you the experience of a conversational AI that lives inside your head that you could talk to like you're talking to yourself internally.

It's like basically your alter ego, if you will, your other mind, in a sense, is the entirety of the internet could eventually be accessible.

as a way of talking to yourself.

So that he finds that very interesting.

So what are some of the the potential uses for this

technology?

So I kind of categorize this into

several buckets here.

So one category of use cases, if you will, would be for disabilities.

So people that have trouble communicating due to conditions like ALS or stroke, cancer, laryngectomy or dysphonia, which I think, is that what RFK Jr.

has?

He has dysphonia.

Yeah, exactly what I said.

that's like the, they said this was their primary focus is to help people with these disabilities.

And I mean, people that can't really talk or communicate like that, they can use a computer in some ways and craft sentences, but

it takes a while to do that.

This is something that could help them communicate so much more simply and easily and quickly.

So I think it could be an incredible boon for a lot of

these sufferers.

But another use case category here is for when, obviously when silence is important.

So when, for example, when talking to somebody in a medical setting could increase infection risk, right?

Because sometimes you just don't want droplets being shared, you know, being launched into the air.

So, the less you talk,

the lower the infection risk.

So, that so this could be used for that specific example.

But also, there's broadcast sets, there's courtrooms, libraries, covert research where silence is important.

That this technology could be very helpful.

Another category is if hearing is obscured for whatever reason.

So, fire rescue teams are trying to coordinate information amongst other rescue team members and there's sirens and other noises on site.

You could communicate this way without worrying about people not hearing you.

But also construction sites, concerts, factories, motorcyclists, all these are all scenarios where hearing could be greatly impaired and this type of communication could be great, greatly

greatly helpful.

Now a lot of this technology, I was surprised that a lot of this technology seems to have already been fairly well well developed, at least according to the demos that I saw.

For at least, like for five or six years, they had the basics of what I've been discussing.

It seemed to me.

So, some of the limitations that they experience, I think, would be something like speaker training.

There's probably a deal, a great deal of speaker training that's probably going to be required if you had this technology right now.

Also, I think that remember when I said that this device will read those signals and map using an AI to map those signals to words.

Well, I found out that they had a 92% accuracy rate, but that was only with a very constrained vocabularies.

So, these weren't really real-word-world testing with people that you have normal vocabularies.

So, 92%, you got to take that with a

grain of

salt.

I mean, yeah, so that's so maybe this is one of the areas where they've improved in the past five or six years.

And there's also reliability issues

that are very likely.

Like if, look, what if you're walking and sweating and what if

you're walking fast and the motion of your body interferes with this device?

And this isn't a subtle device.

This is hanging over your ear, down the side of your face, across the top of your neck, and then up under your lips.

I mean, this is not a subtle thing that you're wearing.

It's not very covert at all.

If you wear this, it's like those Google glasses with the

beeping light,

the flashing light on it.

I'm just curious what improvements have been made in the past half decade on this.

And I think we're going to find out pretty soon.

Kapoor will discuss all through egos technology live this September 17th at Axios AI Summit in Washington, D.C.

So next week, I might be able to have a quick little update on this technology and where it is now,

as opposed to a lot of the information which is kind of out of date,

probably by this point right now.

But it seems interesting and it's got some possibilities for some niche scenarios that could help a a lot of people.

But Steve, so what's your take?

What's your take on this and what I said?

My take is that it looks dorky.

The guy's rolling at the device.

Yeah.

Yeah, it's not subtle.

It's not a terrible thing.

That may be a subtle color.

Yeah.

I mean, sure, it's not implausible that if you're like sub-vocalizing, basically, and it's picking up on your muscle contractions and interpreting that with 90-something percent accuracy, that's not implausible.

But the question is, how transparent are they being about how big the whole thing is and what it is connecting to and what its real functionality is?

I'm just trying to find really solid technical details, and there's just not a lot out there.

Not a lot of that out there.

Hopefully, next week we'll have some more information.

So I'll definitely chime in next week as to what was released

since then.

I don't like them calling it telepathy, though.

That's just

no,

I think that's all like other, you know, I didn't see MIT referring to it it as telepathy at all.

I mean, I have seen like near telepathic

in some places.

So, yeah,

they're not saying that this is telepathy.

And it's not, of course, you know that websites are going to grab onto that idea and couch it in those terms.

I worry about this technology because it sounds like it could really mess people up.

Why?

Because, you know, if you have a direct connection in your brain just by thinking to access the internet and all that stuff, like, come on.

Yeah.

You're just talking about the application of being plugged in.

Yeah.

Plugged in with a deeper level.

How to interact overtly with the world.

Yeah.

Right.

It's definitely a deeper connection than we can experience now.

Hands-free, voice-free.

It's, you know, it's, and it kind of is.

Imagine you think a question and then you hear an answer with no,

you know, external keyboards or voice interactions.

That's got to be a weird feeling.

But I think we need to, we got to like stop moving in the direction of being insular and actually

have communities, and kids got to go out and ride their bikes.

How the hell did we survive those days?

I don't know.

Evan, tell us about these decreasing science scores.

Yeah, sorry.

I wish I had better news here.

The Skeptics Guide #1053 - Sep 13 2025

Listen and follow along

Transcript

More episodes from The Skeptics' Guide to the Universe

The Skeptics Guide #1052 - Sep 6 2025

The Skeptics Guide #1051 - Aug 30 2025

The Skeptics Guide #1050 - Aug 23 2025

The Skeptics Guide #1049 - Aug 16 2025

The Skeptics Guide #1048 - Aug 9 2025