August 30, 2024 • 42 mins
From X to Tesla to SpaceX to xAI, we've got a bunch of stories about various Elon Musk companies and the challenges they face this week. Plus, the MPA takes down a massive media piracy ring, and more!
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:04):
Welcome to tech Stuff, a production from iHeartRadio. Hey there,
and welcome to tech Stuff. I'm your host, Jonathan Strickland.
I'm an executive producer with iHeart Podcasts and how the
tech are you. It's time for the tech news for
the week ending on Friday, August thirtieth, twenty twenty four.
(00:26):
I feel like the entire world is at dragon Con.
I realize that that's just because I live in Atlanta.
That's where dragon CON's a big science fiction fantasy convention
if you're not familiar, And it seems like all of
my friends, like all the people I know, are currently
at dragon Con. So I'm getting a little fomo here.
But I realized a lot of y'all out there also
(00:46):
aren't there, So let us get lost in the warm
embrace of tech news. First up, there's an update on
the Telegram CEO Pavel Durov story. So on Wednesday, I
published an episode about Telegram, the messaging service, and how
French authorities recently detained the CEO, Pavel Durov, after his
(01:10):
private plane landed at an airport outside of Paris. Since then,
a court has indicted Durov on charges that he has
been complicit in numerous criminal activities that have been happening
or facilitated by the Telegram messaging service. So essentially, the
legal argument here is that Telegram has mostly hands off
(01:32):
policy when it comes to content moderation. They're very lax,
they don't do very much content moderation at all, and
that this means Telegram is effectively allowing crimes ranging from
terrorist acts to the spread of child sexual abuse material
see SAM material. I guess that's repetitive. It's like ATM machine. Moreover,
(01:55):
Telegram has a history of denying cooperation with law enforcement
officials who are conducting legal investigations into criminal activity, you know,
essentially like I've got a legal order for a wiretap
and this organization is not being compliant with it. So,
according to the charges, it's not just that Telegram failed
to cooperate with law enforcement, but that legally received requests
(02:20):
were just outright ignored, and that this amounts to the
same thing as assisting criminal activity, thus being complicit. And
France also has a law that states that services that
provide quote unquote cryptology solutions first have to make legal
declarations to France's cybersecurity agency in order to do so.
(02:40):
Legally within France. Now, Telegram does offer end to end
encryption services, but it's only for one on one communications.
Telegram also allows one to many kind of like Twitter
and also chat rooms, so those do not have encryption options,
and even for one on one communication, encryption is not
(03:02):
on by default, you have to turn it on manually.
Durov has already posted the five million euro bail money
that he was hit with, but he has also received
orders not to leave France until this matter has been concluded.
His brother Nikolai, who is also a co founder of Telegram,
also has a warrant out for his arrest in France.
(03:25):
I've got a lot of news relating to Elon Musk
and his various companies. I mean a lot, Like a
lot happened this week, I think just by coincidence. But
let's get through it. First. Up, we have a feud
brewing between a Brazilian Supreme Court judge and Elon Musk,
which I mean, that's a weird sentence, but Musk has
(03:46):
said that this judge, Alexandra de Mores, and I apologize
for butchering pronunciation, but that this judge has threatened to
incarcerate one of X's legal teams stationed in Brazil. If
X did not block some accounts on X that we're
operating within Brazil, those accounts, according to this judge, were
(04:08):
spreading misinformation and hate speech. Murkying up the waters is
the fact that Elon Musk appeared to be very cozy
with kind of a hard right leaning leader of Brazil
who is no longer the leader of Brazil, and that
the Supreme Court judge is kind of an opposition to
(04:30):
that particular political stance. So there's like some political tension
going on here beyond this claim that these accounts are
spreading misinformation. Now, Musk ended up replying to this request
of Hey, you need to shut this stuff down or
I'm going to arrest your lawyers here in Brazil. He
(04:53):
responded by shutting down X's offices in Brazil. He just
closed those offices outright. However, X is still providing service
in Brazil, it just doesn't maintain any offices there. However,
that creates a new legal issue for the company because
in Brazil, if you are providing a messaging company a
(05:15):
messaging service within the nation, you have to have a
presence in Brazil as well. There has to be someone
who's held accountable in the country, and by shutting down
the offices, Musk no longer has that person. So Maraz
could order X to be shut down in Brazil entirely.
And he also ordered that the bank accounts for Starlink
(05:39):
business in Brazil to be frozen. So Starlink is the
satellite internet service provider business. It's a branch of SpaceX,
which of course is another of Elon Musk's companies. And
y'all you know I am not a fan of Elon Musk.
I mean, I don't make that a secret, but it
does seem like a pretty drastic case of overreach if
(06:03):
you ask me, because X and Starlink are not part
of the same company. They may have the same obnoxious
dude behind them, but they aren't the same company. And
to shut down the assets of one company in order
to force a different company to follow your your legal rulings,
(06:25):
that strikes me as being very wrong. I also think
it's important to remember Elon Musk technically he's not the
CEO of X anymore. He hasn't been the CEO of
X for ages. It's Yakarino who's the CEO. But everyone
essentially assumes he's ultimately the one calling the shots over
at X, even if Yakarino is nominally CEO. I hate
(06:49):
to even use that phrase. I'm sure Yakarino does a
lot of work. I have no idea what her day
to day has to be like, nor do I understand
what it must be like to work at a company
where everyone just assumes that you are a figurehead. And
I honestly don't know the extent to which that accusation
could be true. It's a crazy world, Elon Musk's world
(07:11):
in particular. Well, if the judge orders X to get
shut down in Brazil, presumably the way that would work
is he would order Brazilian telecommunications companies to just block
all traffic to X within the country, because clearly the
judge can't force X to shut down operations in Brazil.
I don't even know how X would do that to
(07:33):
begin with, but he could force the various telecommunications companies
to block traffic to X. However, people in Brazil could
potentially still get to X by using a VPN service
that was based outside of the country. Anyway, X and
Starlink have both accused the judge of breaking Brazilian law
(07:56):
and say they will escalate legal proceedings within the country.
I have no idea how that works, because at least
here in the United States, once you get up to
the Supreme Court level, there's nowhere else to go. That's
that's the top. But it definitely seems to me like
the judge is going a little judge dread over there.
So I'm sure there's a lot more complex issues at
(08:20):
play in this story. Maybe I'll do a full episode
about it in the future, because I think it does
require a nuanced approach to kind of understand everything that's
going on. I'm sure it's not a black and white
situation for either side, But I don't know that freezing
of assets for starlink, it just strikes me as like,
(08:41):
how could there be a legal basis for that. I
guess if you argue that there was like a single
parent company and all these other companies were underneath that,
maybe then, but I don't think that's how the corporate
structure works between SpaceX and X apart from the fact
that they share Elon Musk's you know, obsession with the
(09:03):
letter X. Meanwhile, over here in the United States, X,
formerly known as Twitter, was kind of a caught out
when it marked links to an NPR news story as
being unsafe. So, yeah, X will not block accounts in
Brazil because that's a free speech issue, but here in
(09:24):
the US, it will label NPR stories as being unsafe.
So if you try to find out what the deal
is between Donald Trump's staff, campaign staff and an employee
of Arlington National Cemetery by following a link to an
NPR story on the matter, then you're told, hey, you
don't want to do that. This is taking you to
a bad place. The unsafe marker is typically something that
(09:47):
pops up if the link is to a site that's
known to host like militious software or sometimes things like
outright misinformation. Now, NPR might occasionally play some truly terrible
jazz on the radio, but I wouldn't go so far
as to call it malicious. The warning was in the
(10:07):
form of a pop up as well. It wasn't just
like a little label. If you clicked on the link,
you would get a pop up and it would say
that this link might be unsafe, and then there'd be
some text, and underneath that would be a big old
blue button that if you clicked on, it would just
take you right back to X, so it would, you know,
derail you from going to NPR. Beneath that big old
(10:31):
blue button is a smaller line of text with a
hyperlink that would let you continue on to read the
news story. So that is kind of crazy right now.
This could just be a simple mistake, but Musk has
already shown contempt for NPR because previously he labeled NPR
(10:52):
as a state affiliated media outlet. Typically that label indicates
a media company that is controlled by a governmental body,
like it's a propaganda arm. In other words, that is
not what NPR is. NBR receives a little bit of
federal funding, but only a tiny little bit. Most of
(11:13):
its funding does not come from federal sources, so and
there is no federal oversight or control of what goes
out over NPR. So this was a very misleading label
that Musk applied to NPR. Following that NPR chose to
leave X. So, you know, Musk has also over the
(11:36):
last few years been promoting more right leaning ideologies and
personalities on X, so I think it would be very
easy to view this and see it as a purposeful
attempt to discourage X users from going to NPR to
read up about this incident that happened at Arlington National Cemetery. However,
(11:59):
X state that the labeling was actually a false positive,
that this was not intentional, it was an honest mistake.
Maybe that's true, Like, it's easy for me to believe
that it was an intentional move to try and dissuade
people from reading up about what happened at Arlington National Cemetery.
But it's also very much true that over the last
(12:21):
year X has has eliminated so many people from the
company payroll that mistakes like this could easily happen simply
because they don't have the people to make sure it
doesn't happen. So either version could be true. Okay, so
we've got stories about X and Starlink. What about Tesla, Well,
(12:44):
there's more bad news on that, So Tesla, of course
yet another elon Musk company, and Christian Agatti of autoevolution
dot Com has an article titled third documented Tesla cybertruck
fire in less than a month raises questions woof so
(13:04):
a gad points out that in the marketing push for
cyber truck, Tesla was really stressing how the cyber truck
is a resilient vehicle right now. Granted, in the midst
of trying to say the cyber truck could withstand tons
of abuse, they accidentally broke some of the cyber truck's
windows while trying to show how tough they are. That
(13:25):
was a whoopsie, but anyway, the most recent of these
fires happened after an accident in which a cyber truck
driver hit a fire hydrant. My guess is that then
caused the battery inside the cyber truck to experience a
short where there was damage to the battery, and that
(13:47):
that created a short circuit that led to overheating and
then fire. That's my guess. I don't know for sure
that that is what happened. There has yet to have
been an investigation, or at least results of an investigation
shared publicly, so maybe it turns out something else caused
the fire. This happened in Harlingen, Texas. Don't know if
(14:09):
that's how you say it. There are a lot of
places in Texas where I think it's pronounced one way.
I am totally wrong. Earlier in the month, however, news
broke about a different incident that happened in Bayton area,
which is also in Texas, and In that one, a
cyber truck ran into like a concrete ditch trench or
(14:29):
something and caught fire. Tragically, there was a fatality, although
last I saw, they had not yet determined what actually
caused the death, whether it was a medical emergency, the
accident itself, or the fire. The third fire incident that
was one where Agady reported that a colleague of his
(14:49):
had actually found a burnt cyber truck listed on an
online auction site, and the description indicated that the vehicle
had been in a wreck and in fact subsequently caught fire,
and that this happened in Houston, Texas. So three accidents
resulting in fires, all three happening in Texas in the
span of a month. Now, I think it's too early
(15:12):
to draw any firm conclusions as to whether this means
the cyber truck actually is a flawed vehicle in that regard,
you know, like if the panels protecting the electric battery
aren't sufficient or whatever. I just don't know that that's
the case. It could be, but I think we need
to have an actual investigation to determine that. So I
(15:32):
think that's probably going to happen. I mean, with three
fires in a month, it seems to me like it's
at least warranted to take a look and see if
there are any common factors here at play. Maybe there aren't.
Maybe this is just coincidence and it all happened around
the same state and within the same month. That's possible,
(15:53):
so we need to investigate to find out. Okay, we
got lots more news to get through. Let's take a
quick break to thank our sponsors. We're back, and now
we're not done with Elon Musk news items just yet.
(16:18):
So over at SpaceX, there was a different kind of setback. Now,
while I'm talking about space and SpaceX, I should also
give a quick update and mention that we have this
ongoing issue of the two astronauts who went up on
the Boeing star Liner that then docked with the International
(16:39):
Space Station for a couple of months. Now. NASA was
hoping to find a way where they could return the
astronauts to Earth aboard the star Liner spacecraft. That was
something that Boeing was saying would be within acceptable risk parameters,
but NASA engineers ultimately disagreed with that. So the official
(16:59):
decision is that the Starliner is going to return to
Earth empty and the two astronauts will return home early
next year aboard a SpaceX Dragon capsule. Now, the SpaceX
story I wanted to touch on for this week is
that on Wednesday of this week, a Falcon nine first
stage booster had a big ol' whoopsie when it returned
(17:21):
to land on an ocean platform. Now, as I'm sure
you're all aware, one of the big draws for SpaceX's
launch vehicles is that they are reusable. This cuts way
down on the cost of launch, which is a huge
limiting factor for space The first stage booster landed on
this ocean platform, but then shortly thereafter it fell over
(17:44):
and it burst into flames. Fortunately, there were no reports
of any injuries or anything like that. There were no
reports of any public damage from this particular incident. However,
the FAA has grounded the Falcon nine launch vehicles. SpaceX
can investigate the incident and then share its findings and
(18:04):
then after any corrective actions have been ordered as a
result of that investigation. That means no SpaceX launches until
the FAA lifts that order. And that also means there
was supposed to be a launch I think today actually
that was going to deliver more Starlink satellites into orbit,
and that has since been delayed. Actually, the booster that
(18:26):
burst into flames, that was part of a mission of
sending more Starlink satellites into orbit as well. But the
second one that was supposed to happen now has been
postponed indefinitely until the ban has been lifted once they
figure out exactly what happened with this accident. No telling
yet if this is going to impact other upcoming launches.
(18:48):
There are a couple that are coming up before too long,
But it may be that this whole thing is resolved
relatively quickly. We'll just have to keep an eye out.
The hits keep on coming though. From us. In Memphis, Tennessee,
the Southern Environmental Law Center alleges that Musk's AI company,
which of course is called Xai, is using natural gas
(19:11):
burning turbines to power the AI facilities. We all know
that these AI data centers require a huge amount of
electricity in order to work. So one of the things
that companies that are using these kinds of power sources
that like natural gas burning turbines, one of the things
(19:32):
they're supposed to do is file a whole bunch of
paperwork to get the right permits to actually operate that
kind of machinery, and according to the Southern Environmental Law Center,
they say that there's no record that Xai actually followed
that process in Tennessee. So in an open letter, the
(19:53):
organization claims that Xai has twenty gas turbines operating without
first applying for those permits and has called upon the
local health department to investigate Xai to determine if, in fact,
that is what is going on now. I can't say
I would be surprised to hear that a Musk backed
company jumped into a situation like that without proper paperwork,
(20:15):
but I have to be fair. That is my bias talking,
Like just because Musk has proven that, you know, he's
done that kind of thing before, it doesn't mean that
that's what's happened this time. It could all be totally
a misunderstanding, like maybe the paperwork has been filed and
they just didn't find record of it. But they are
calling for an investigation to find out if the company
(20:37):
in fact followed the right process in order to operate
this way, because the concern here is that the operation
of these twenty gas turbines could contribute to pollution in
the Memphis area. Meanwhile, the Center for Countering Digital Hate
released the results of a study focusing on the GROCK
AI assistant that's the AI assistant connected to YEP, formerly
(21:00):
known as Twitter. YadA YadA, YadA so. This study concluded
that GROC lacks sufficient guardrails to prevent people from using
it to generate misleading images of real public personas, including politicians.
The study said that the tool was able to create
a convincing fake image of Donald Trump appearing to be
very sick in a hospital bed, and another of Kamala
(21:24):
Harris appearing to do drugs. The researcher said they tried
various methods to generate these images that range from just
blatantly giving a text prompt asking for what they wanted
to see, to taking a more circumspect route where they
used descriptive language to get the same effect, to get
the same results they wanted without actually triggering any preventative
(21:47):
safety features that otherwise would stop that kind of request
from working. I find this particularly frustrating, not just because
of the disinformation aspect, which is clearly bad enough alone,
but also because Elon Musk like again, I don't like
the guy, but he has repeatedly sounded the alarm about
(22:08):
the dangers of AI, and I agree AI is potentially
very dangerous if it's implemented poorly. And he has argued
that AI needs to be developed and deployed responsibly. I
agree with that too. I think that AI does need
to be developed and deployed responsibly, and that there should
be plentiful safety measures in place to prevent AI from
causing harm. I agree with that. All of these things
(22:31):
I agree with, And yet here is an AI tool
on one of Elon Musk's own platforms that clearly fails
to follow any of those directives. But then you could
also make the same argument that the AI found in
Tesla vehicles has at least appeared to be less reliable
than the company tends to market to potential customers. So
(22:54):
there's that as well. There's also been a recent change
to Grock, the AI assistant on it, apparently in response
to a complaint letter sent to the company from the
offices of five different Secretaries of State, so those states
would be Minnesota, New Mexico, Washington, Pennsylvania, and Michigan. The
(23:14):
offices were complaining that Groc was giving incorrect information in
response to prompts asking about ballot deadlines in those states.
So essentially this is related to election information, and GROC
apparently was just giving incorrect responses when people were asking
these basic questions about ballots. So now rather than just
(23:37):
giving an incorrect answer, GROC will respond to those kinds
of requests by urging the visitor to go to vote
dot gov for the actual answer, So that I think
is a market improvement. Tom David S. Pumpkins Hanks has
a message for all of y'all out there, and that
(23:57):
message is to not trust drug advertisements that seem to
have Tom Hanks voicing the ad copy. So specifically, on Instagram,
Hanks posted quote, there are multiple ads over the internet
falsely using my name, likeness, and voice promoting miracle cures
and wonder drugs. These ads have been created without my consent,
(24:21):
fraudulently and through AI. I have nothing to do with
these posts, or the productions and treatments, or the spokespeople
touting these cures. I have Type two diabetes and I
only work with my board certified doctor regarding my treatment.
Do not be fooled, Do not be swindled, Do not
(24:42):
lose your hard earned money. End quote. Now, if all
this sounds a little familiar, that might be because of
something that happened around the same time last year, where
there was this ad video going around with an AI
generated deep fake version of Tom Hanks prom voting a
dental plan Lisa needs braces dental plan. Sorry, sorry, sorry.
(25:06):
Some episodes of The Simpsons are kind of hardwired into
my brain and I just go into like a feedback loop.
I have to admit that when I was thinking about
how deep fakes were going to spread beyond the creation
of pornographic images and video, which is its own thing
and it's horrifying and terrible, when I was thinking about
how it was going to branch out beyond that, I
(25:27):
had not clocked Tom Hanks as being one of the
first high profile victims. But here we are. Bloomberg reported
that Apple is cutting one hundred positions in its digital
services department, and that the affected employees will have two
months to find another position within the company or else
be struck from the roster entirely. So apparently the main
(25:49):
group affected is the books team, which, as Samuel Aksin
of Ours Technica has pointed out, has no subscription service
attached to that particular product. So it sounds like Apple
really wants to redirect energy towards services that more directly
impact revenue. Which I mean, you can understand that from
(26:10):
a business perspective, right, Like, why would you focus on
things that don't impact revenue? Maybe they impact the value
or the perception of value, which you could argue is
a reasonable reason to keep it going. But yeah, most
companies are going to say revenue is way more important.
(26:32):
So it appears that Apple is one of those. I
hope anyone who has been affected by this will find
a new position with a great team as soon as
is possible. It's been a pretty long time since I've
talked about Uber getting hit with fines, but this next
story is quite a doozy. In fact, according to ngadget's
(26:52):
Sarah Fielding, it would be the largest fine Uber has
ever faced, a whopping two hundred and ninety million Euro penalty.
That's the equivalent of around three hundred and twenty four
million dollars. So what the heck happened? Well, this has
to do with the EU and GDPR. GDPR that's the
(27:13):
set of rules that are meant to protect EU citizens'
personal data. So one thing that companies are not supposed
to do under these rules is to send the personal
information belonging to EU citizens outside the EU itself unless
you're using appropriate measures to make sure you're protecting their information.
(27:34):
This is a thing that a lot of platforms have
run into where they were found to be transmitting information
to servers in the United States, but the information belonged
to EU citizens, and without those protections in place, these
companies can't guarantee the safety of that information. Because GDPR
is so focused on the safety of personal information, that's
(27:57):
a problem. So the Dutch Data Protection Authority or Dutch
DPA says that Uber really failed to protect information for
Uber drivers and that included stuff like their taxi licenses,
their identity documents and lots more other information. That Uber
transmitted this info overseas to Uber data servers in the
(28:20):
United States while not using any approved transferred tools that
are meant to ensure the safety of those data transfers.
The Dutch DPA said this failure to provide adequate protection
is a violation and that Uber subsequently has to pay up.
And it's notable here that as far as I can tell,
no unauthorized individuals ever got access to this information. This
(28:43):
isn't like a data breach or something, so it's not that,
you know, Uber failed to protect the info and someone
got hold of it. It's that Uber was not following
the approved process to make sure data was handled appropriately,
so arguably that personal information could have been intercepted or
exploited by other parties. And Uber reps have said that
(29:04):
the company will of course appeal the fine, so we'll
have to see where this develops from here. Okay, I've
got some more news stories to talk about. Before we
get to those, let's take another quick break. Okay, let's
(29:25):
wrap up some of our tech news this week over
on blue Sky, which, in case you're not familiar with
blue Sky, it's an alternative to Twitter, slash x, along
with other similar tools like Mastodon or threads. So blue
Sky has added some new anti toxicity features and I
think the features are interesting. I'm not sure that it's
(29:49):
the best approach, but they are interesting. It's hard to
determine how to best implement anti toxicity measures on a
platform that allows people to just join up. I mean,
as platforms get larger, they become more difficult to moderate,
assuming that you have a platform that's interested in moderation
in the first place. So yeah, I see also Telegram.
(30:11):
But the big one that I wanted to talk about
was that users are able now to remove their original
post if someone else quotes that post. So in other words,
let's say, let's say that you post something that says like,
I really enjoyed the Barbie movie, which is awesome. Honestly,
(30:31):
I enjoyed the Barbie movie. But let's say that I'm
not me. Let's say that I am the Evil Mirror
Universe version of me, and I've decided, because you have
claimed that you enjoyed the Barbie movie, you are deserving
of criticism and trolling and all that kind of stuff.
So I then quote your post, and I then put
(30:55):
some really nasty commentary underneath about how I think you're
a dumb dumb for liking a perfectly good movie. And
now you can actually take action on this, so before
you might just end up being flooded with like toxic notifications,
let's say that Evil Mirror Universe me has a lot
(31:15):
more notoriety and a lot more followers on Blue Sky.
So I post this negative takedown of your post and
everyone dogpiles on top of you. Now on Blue Sky,
you could go in and remove the quoted tweet from
my post. So you know, if my post is centered
(31:37):
around your earlier post, and you remove your earlier post
from mine, then all it's left is the commentary. If
that commentary is not specific, people don't even know who
I'm referring to as I rant about someone daring to
like the Barbie movie. So that is effective now, granted,
if in my narrative, if in my common Harry I
(32:00):
have also identified you, then removing your post from the
quotation doesn't really help that much because people can still
see who I'm talking about. Maybe they don't get a
link to your profile or whatever, but there's limited utility here.
But yeah, I think that that's generally a positive step.
(32:21):
I'm not sure that's the most effective one, but it
is meant to prevent instances of dog piling. So that
if someone takes your post and then, out of context,
tries to argue that you're a terrible person or whatever,
you can just remove the quotation part out of that post.
Now you can't touch anything else that's in that post, right, Like,
(32:43):
if in my description, I really drag you. You can't
go in and change my description, like you can't change
the commentary I have added. You can just remove the
quoted part, which I think makes sense, Like you shouldn't
have the ability to be able to change what someone's
post reads, because then it's just chaos. But I think
(33:05):
it is a good step. I just think it's tough,
like this is a hard problem to fix. Another similar
feature on blue Sky is that you can actually hide
replies to your posts. So let's say you post something
and no one's quoting you, but they are replying to you,
and some of them are being real nasty. You can
actually hide those replies, and you can choose to hide
(33:25):
it just from yourself, which is like blocking or muting someone,
or you can choose to hide that reply from everybody
so no one can see that this person is just
being nasty towards you. So when someone's being hateful, you
can hide so that no one sees it. Or maybe
you just shield yourself and you let it stand for
everybody else. Blue Sky says it will continue to introduce
(33:48):
features to make the platform a more safe and welcoming place.
This I think is in stark contrast to X like
over at X. You don't really see a Musk and
company putting in measures to make X a safer and
more welcoming place. You do occasionally see stories like the
(34:08):
one I mentioned about NPR, where it appears that whenever
they do enter into content moderation, it's to suppress any
any kind of viewpoint that doesn't go into alignment with
what Musk believes. That at least that's the appearance. Whether
that's what's really going on or not, I don't know.
Like I said, it could really be that there was
(34:30):
just an error in a false positive, But yeah, Blue
Sky strikes me as taking kind of like a maybe
not a total one to eighty approach from X, but
a substantial deviation from that approach. The Motion Picture Association
announced this week that through partnering with authorities in Vietnam,
they have shut down an online media piracy ring that
(34:53):
had accumulated billions of visits between them from January twenty
twenty three up to June twenty twenty four. So sort
of the hub of this, like the central crux was
F Movies, and under F Movies, there were several piracy
sites that were in operation. They included any Wave Flicks
(35:15):
Tours my Flixer and others like that. So the MPa
referred to f Movies as the mothership of piracy and
that the takedown had a cascading effect across these various
piracy sites which have been hosting media illegally for years
and distributing it to whomever goes there. Various media companies
(35:39):
and organizations that represent media companies have spent decades fighting
online piracy, often in ways that end up being outlandishly disproportionate.
Like you might remember, I mean I remember because I
remember the nineties when the RIAA really began to shake
down like service pro like Verizon to hand over the
(36:02):
identities of customers that the RIAA suspected were guilty of piracy.
There are people who were hit with truly ridiculous, astronomic
fines for piracy. And then there's the various issue of
media organizations arguing that piracy equates to loss of revenue,
(36:26):
like if you watch a movie illegally, if you download
a music track illegally, that means we lost revenue. A
lot of entities, including the United States Government Accountability Office,
say that argument doesn't hold water. At best, it is misleading.
At worst, it is wrong, and that's because it's impossible
(36:47):
to determine if someone who pirated a piece of media
otherwise would have purchased that piece of media legally. Let's
say that piracy was impossible somehow, Like we're in a
world where it's just impossible to pirate stuff. Someone who
would have pirated things, you can't say that they will,
though otherwise go out and buy it. Maybe they just
go without. You can't make the argument that a downloaded
(37:10):
piece of media equates to loss of revenue, So a
lost sale isn't the same thing as stolen money. Like
if I walk up to a movie theater and I
look at the list of films that are available, and
I ultimately decide, eh, I don't feel like watching any
of these, and I walk off. I didn't steal from
(37:30):
any of the movie companies. So yeah, I think piracy
is bad. Just to be clear, I do think piracy
is bad. I do not think you should do it.
Don't be a pirate. However, I think media companies that
take the nuclear option based on unsupportable arguments are bad too.
No one's in the right in that particular situation. But
(37:53):
you can read more about this particular story about the
Motion Picture Association and f movies both On Technica and
The Verge. Both sites have some great articles about this
recent Hollywood war against online piracy. I recommend you go
check those out now. For other recommended reading this week,
I've got a couple of different pieces I want to
talk about. First up is a paper that was published
(38:16):
in Nature, the science journal Nature. This paper is by
Valentine Hoffmann, Pratyusha rha Kaluri, Dan Jorowski, and Sharise King.
It is titled AI generates covertly racist decisions about people
based on their dialect. So, as you might suspect, this
(38:37):
paper investigates the issue of bias found within AI models,
but this time with regard to dialect as opposed to say,
physical features. I think it's a really important piece, particularly
since we live in a world in which code switching
is a really important mechanism for navigating different cultural groups
for a lot of people, and if they are switched
(38:57):
into one particular culture, it's entirely possible that an AI
model is going to make some very racist conclusions based
on that anyway. Next up would be a piece by
Heather Knight and Kate Conger of The New York Times
titled Elon Musk's X is leaving San Francisco. City officials
(39:19):
say good riddance. So this story covers how X, which
has long had its HQ in San Francisco, California, is
now moving to other parts of California, like it's going
to rely on offices that are in other cities in California,
but the HQ itself is relocating to Texas, And how
some people who are in various positions in San Francisco
(39:41):
government aren't exactly wringing their hands at x's departure. Then finally,
there's a concerning article in The Texas Observer by Francesca
du Nunzio titled Texas State Police gear up for massive
expansion of surveillance tech. It serves as a reminder that
it's very important to hold law enforcement agencies accountable and
(40:03):
to investigate whether their reliance on certain technologies is merited,
particularly since some of those technologies have proven to have
flaws that can lead to disproportionate harm being done to
vulnerable populations. Specifically, I'm thinking about things like facial recognition
technologies and how that often can disproportionately affect innocent people
(40:23):
from the black community. We've seen that happen numerous times,
So any surveillance state stuff always makes me nervous. It's
also wild that Texas identity is so strange to me
because there's kind of ingrained in Texas is this idea
of independence and being able to stand on your own,
(40:44):
to hike yourself up by your bootstraps, and that you know,
like a lot of the narrative around Texas government is
about being self sufficient. I mean the state of Texas.
Their power grid is independent of the national power grid
in the United States, for example. So you've got this
idea of individualism and self sufficiency, and on the flip side,
(41:07):
you've got this growing surveillance state in law enforcement, and
the two don't really gel together to me, apart from
a slow moved more toward authoritarianism, which I think is
pretty concerning you. But the tech enabling that is really
the feature of that particular article, So check that out.
(41:30):
I hope all of you have a great weekend here
in the United States, the holiday weekend. I hope you
have a great holiday if you're here in the US.
If you're not in the US, I hope you just
have a great weekend period. And I'll talk to you
again really soon. Tech Stuff is an iHeartRadio production. For
(41:53):
more podcasts from iHeartRadio, visit the iHeartRadio app, Apple Podcasts,
or wherever you listen to your favorite show still
Welcome to tech Stuff, a production from iHeartRadio. Hey there,
and welcome to tech Stuff. I'm your host, Jonathan Strickland.
I'm an executive producer with iHeart Podcasts and how the
tech are you. It's time for the tech news for
the week ending on Friday, August thirtieth, twenty twenty four.
(00:26):
I feel like the entire world is at dragon Con.
I realize that that's just because I live in Atlanta.
That's where dragon CON's a big science fiction fantasy convention
if you're not familiar, And it seems like all of
my friends, like all the people I know, are currently
at dragon Con. So I'm getting a little fomo here.
But I realized a lot of y'all out there also
(00:46):
aren't there, So let us get lost in the warm
embrace of tech news. First up, there's an update on
the Telegram CEO Pavel Durov story. So on Wednesday, I
published an episode about Telegram, the messaging service, and how
French authorities recently detained the CEO, Pavel Durov, after his
(01:10):
private plane landed at an airport outside of Paris. Since then,
a court has indicted Durov on charges that he has
been complicit in numerous criminal activities that have been happening
or facilitated by the Telegram messaging service. So essentially, the
legal argument here is that Telegram has mostly hands off
(01:32):
policy when it comes to content moderation. They're very lax,
they don't do very much content moderation at all, and
that this means Telegram is effectively allowing crimes ranging from
terrorist acts to the spread of child sexual abuse material
see SAM material. I guess that's repetitive. It's like ATM machine. Moreover,
(01:55):
Telegram has a history of denying cooperation with law enforcement
officials who are conducting legal investigations into criminal activity, you know,
essentially like I've got a legal order for a wiretap
and this organization is not being compliant with it. So,
according to the charges, it's not just that Telegram failed
to cooperate with law enforcement, but that legally received requests
(02:20):
were just outright ignored, and that this amounts to the
same thing as assisting criminal activity, thus being complicit. And
France also has a law that states that services that
provide quote unquote cryptology solutions first have to make legal
declarations to France's cybersecurity agency in order to do so.
(02:40):
Legally within France. Now, Telegram does offer end to end
encryption services, but it's only for one on one communications.
Telegram also allows one to many kind of like Twitter
and also chat rooms, so those do not have encryption options,
and even for one on one communication, encryption is not
(03:02):
on by default, you have to turn it on manually.
Durov has already posted the five million euro bail money
that he was hit with, but he has also received
orders not to leave France until this matter has been concluded.
His brother Nikolai, who is also a co founder of Telegram,
also has a warrant out for his arrest in France.
(03:25):
I've got a lot of news relating to Elon Musk
and his various companies. I mean a lot, Like a
lot happened this week, I think just by coincidence. But
let's get through it. First. Up, we have a feud
brewing between a Brazilian Supreme Court judge and Elon Musk,
which I mean, that's a weird sentence, but Musk has
(03:46):
said that this judge, Alexandra de Mores, and I apologize
for butchering pronunciation, but that this judge has threatened to
incarcerate one of X's legal teams stationed in Brazil. If
X did not block some accounts on X that we're
operating within Brazil, those accounts, according to this judge, were
(04:08):
spreading misinformation and hate speech. Murkying up the waters is
the fact that Elon Musk appeared to be very cozy
with kind of a hard right leaning leader of Brazil
who is no longer the leader of Brazil, and that
the Supreme Court judge is kind of an opposition to
(04:30):
that particular political stance. So there's like some political tension
going on here beyond this claim that these accounts are
spreading misinformation. Now, Musk ended up replying to this request
of Hey, you need to shut this stuff down or
I'm going to arrest your lawyers here in Brazil. He
(04:53):
responded by shutting down X's offices in Brazil. He just
closed those offices outright. However, X is still providing service
in Brazil, it just doesn't maintain any offices there. However,
that creates a new legal issue for the company because
in Brazil, if you are providing a messaging company a
(05:15):
messaging service within the nation, you have to have a
presence in Brazil as well. There has to be someone
who's held accountable in the country, and by shutting down
the offices, Musk no longer has that person. So Maraz
could order X to be shut down in Brazil entirely.
And he also ordered that the bank accounts for Starlink
(05:39):
business in Brazil to be frozen. So Starlink is the
satellite internet service provider business. It's a branch of SpaceX,
which of course is another of Elon Musk's companies. And
y'all you know I am not a fan of Elon Musk.
I mean, I don't make that a secret, but it
does seem like a pretty drastic case of overreach if
(06:03):
you ask me, because X and Starlink are not part
of the same company. They may have the same obnoxious
dude behind them, but they aren't the same company. And
to shut down the assets of one company in order
to force a different company to follow your your legal rulings,
(06:25):
that strikes me as being very wrong. I also think
it's important to remember Elon Musk technically he's not the
CEO of X anymore. He hasn't been the CEO of
X for ages. It's Yakarino who's the CEO. But everyone
essentially assumes he's ultimately the one calling the shots over
at X, even if Yakarino is nominally CEO. I hate
(06:49):
to even use that phrase. I'm sure Yakarino does a
lot of work. I have no idea what her day
to day has to be like, nor do I understand
what it must be like to work at a company
where everyone just assumes that you are a figurehead. And
I honestly don't know the extent to which that accusation
could be true. It's a crazy world, Elon Musk's world
(07:11):
in particular. Well, if the judge orders X to get
shut down in Brazil, presumably the way that would work
is he would order Brazilian telecommunications companies to just block
all traffic to X within the country, because clearly the
judge can't force X to shut down operations in Brazil.
I don't even know how X would do that to
(07:33):
begin with, but he could force the various telecommunications companies
to block traffic to X. However, people in Brazil could
potentially still get to X by using a VPN service
that was based outside of the country. Anyway, X and
Starlink have both accused the judge of breaking Brazilian law
(07:56):
and say they will escalate legal proceedings within the country.
I have no idea how that works, because at least
here in the United States, once you get up to
the Supreme Court level, there's nowhere else to go. That's
that's the top. But it definitely seems to me like
the judge is going a little judge dread over there.
So I'm sure there's a lot more complex issues at
(08:20):
play in this story. Maybe I'll do a full episode
about it in the future, because I think it does
require a nuanced approach to kind of understand everything that's
going on. I'm sure it's not a black and white
situation for either side, But I don't know that freezing
of assets for starlink, it just strikes me as like,
(08:41):
how could there be a legal basis for that. I
guess if you argue that there was like a single
parent company and all these other companies were underneath that,
maybe then, but I don't think that's how the corporate
structure works between SpaceX and X apart from the fact
that they share Elon Musk's you know, obsession with the
(09:03):
letter X. Meanwhile, over here in the United States, X,
formerly known as Twitter, was kind of a caught out
when it marked links to an NPR news story as
being unsafe. So, yeah, X will not block accounts in
Brazil because that's a free speech issue, but here in
(09:24):
the US, it will label NPR stories as being unsafe.
So if you try to find out what the deal
is between Donald Trump's staff, campaign staff and an employee
of Arlington National Cemetery by following a link to an
NPR story on the matter, then you're told, hey, you
don't want to do that. This is taking you to
a bad place. The unsafe marker is typically something that
(09:47):
pops up if the link is to a site that's
known to host like militious software or sometimes things like
outright misinformation. Now, NPR might occasionally play some truly terrible
jazz on the radio, but I wouldn't go so far
as to call it malicious. The warning was in the
(10:07):
form of a pop up as well. It wasn't just
like a little label. If you clicked on the link,
you would get a pop up and it would say
that this link might be unsafe, and then there'd be
some text, and underneath that would be a big old
blue button that if you clicked on, it would just
take you right back to X, so it would, you know,
derail you from going to NPR. Beneath that big old
(10:31):
blue button is a smaller line of text with a
hyperlink that would let you continue on to read the
news story. So that is kind of crazy right now.
This could just be a simple mistake, but Musk has
already shown contempt for NPR because previously he labeled NPR
(10:52):
as a state affiliated media outlet. Typically that label indicates
a media company that is controlled by a governmental body,
like it's a propaganda arm. In other words, that is
not what NPR is. NBR receives a little bit of
federal funding, but only a tiny little bit. Most of
(11:13):
its funding does not come from federal sources, so and
there is no federal oversight or control of what goes
out over NPR. So this was a very misleading label
that Musk applied to NPR. Following that NPR chose to
leave X. So, you know, Musk has also over the
(11:36):
last few years been promoting more right leaning ideologies and
personalities on X, so I think it would be very
easy to view this and see it as a purposeful
attempt to discourage X users from going to NPR to
read up about this incident that happened at Arlington National Cemetery. However,
(11:59):
X state that the labeling was actually a false positive,
that this was not intentional, it was an honest mistake.
Maybe that's true, Like, it's easy for me to believe
that it was an intentional move to try and dissuade
people from reading up about what happened at Arlington National Cemetery.
But it's also very much true that over the last
(12:21):
year X has has eliminated so many people from the
company payroll that mistakes like this could easily happen simply
because they don't have the people to make sure it
doesn't happen. So either version could be true. Okay, so
we've got stories about X and Starlink. What about Tesla, Well,
(12:44):
there's more bad news on that, So Tesla, of course
yet another elon Musk company, and Christian Agatti of autoevolution
dot Com has an article titled third documented Tesla cybertruck
fire in less than a month raises questions woof so
(13:04):
a gad points out that in the marketing push for
cyber truck, Tesla was really stressing how the cyber truck
is a resilient vehicle right now. Granted, in the midst
of trying to say the cyber truck could withstand tons
of abuse, they accidentally broke some of the cyber truck's
windows while trying to show how tough they are. That
(13:25):
was a whoopsie, but anyway, the most recent of these
fires happened after an accident in which a cyber truck
driver hit a fire hydrant. My guess is that then
caused the battery inside the cyber truck to experience a
short where there was damage to the battery, and that
(13:47):
that created a short circuit that led to overheating and
then fire. That's my guess. I don't know for sure
that that is what happened. There has yet to have
been an investigation, or at least results of an investigation
shared publicly, so maybe it turns out something else caused
the fire. This happened in Harlingen, Texas. Don't know if
(14:09):
that's how you say it. There are a lot of
places in Texas where I think it's pronounced one way.
I am totally wrong. Earlier in the month, however, news
broke about a different incident that happened in Bayton area,
which is also in Texas, and In that one, a
cyber truck ran into like a concrete ditch trench or
(14:29):
something and caught fire. Tragically, there was a fatality, although
last I saw, they had not yet determined what actually
caused the death, whether it was a medical emergency, the
accident itself, or the fire. The third fire incident that
was one where Agady reported that a colleague of his
(14:49):
had actually found a burnt cyber truck listed on an
online auction site, and the description indicated that the vehicle
had been in a wreck and in fact subsequently caught fire,
and that this happened in Houston, Texas. So three accidents
resulting in fires, all three happening in Texas in the
span of a month. Now, I think it's too early
(15:12):
to draw any firm conclusions as to whether this means
the cyber truck actually is a flawed vehicle in that regard,
you know, like if the panels protecting the electric battery
aren't sufficient or whatever. I just don't know that that's
the case. It could be, but I think we need
to have an actual investigation to determine that. So I
(15:32):
think that's probably going to happen. I mean, with three
fires in a month, it seems to me like it's
at least warranted to take a look and see if
there are any common factors here at play. Maybe there aren't.
Maybe this is just coincidence and it all happened around
the same state and within the same month. That's possible,
(15:53):
so we need to investigate to find out. Okay, we
got lots more news to get through. Let's take a
quick break to thank our sponsors. We're back, and now
we're not done with Elon Musk news items just yet.
(16:18):
So over at SpaceX, there was a different kind of setback. Now,
while I'm talking about space and SpaceX, I should also
give a quick update and mention that we have this
ongoing issue of the two astronauts who went up on
the Boeing star Liner that then docked with the International
(16:39):
Space Station for a couple of months. Now. NASA was
hoping to find a way where they could return the
astronauts to Earth aboard the star Liner spacecraft. That was
something that Boeing was saying would be within acceptable risk parameters,
but NASA engineers ultimately disagreed with that. So the official
(16:59):
decision is that the Starliner is going to return to
Earth empty and the two astronauts will return home early
next year aboard a SpaceX Dragon capsule. Now, the SpaceX
story I wanted to touch on for this week is
that on Wednesday of this week, a Falcon nine first
stage booster had a big ol' whoopsie when it returned
(17:21):
to land on an ocean platform. Now, as I'm sure
you're all aware, one of the big draws for SpaceX's
launch vehicles is that they are reusable. This cuts way
down on the cost of launch, which is a huge
limiting factor for space The first stage booster landed on
this ocean platform, but then shortly thereafter it fell over
(17:44):
and it burst into flames. Fortunately, there were no reports
of any injuries or anything like that. There were no
reports of any public damage from this particular incident. However,
the FAA has grounded the Falcon nine launch vehicles. SpaceX
can investigate the incident and then share its findings and
(18:04):
then after any corrective actions have been ordered as a
result of that investigation. That means no SpaceX launches until
the FAA lifts that order. And that also means there
was supposed to be a launch I think today actually
that was going to deliver more Starlink satellites into orbit,
and that has since been delayed. Actually, the booster that
(18:26):
burst into flames, that was part of a mission of
sending more Starlink satellites into orbit as well. But the
second one that was supposed to happen now has been
postponed indefinitely until the ban has been lifted once they
figure out exactly what happened with this accident. No telling
yet if this is going to impact other upcoming launches.
(18:48):
There are a couple that are coming up before too long,
But it may be that this whole thing is resolved
relatively quickly. We'll just have to keep an eye out.
The hits keep on coming though. From us. In Memphis, Tennessee,
the Southern Environmental Law Center alleges that Musk's AI company,
which of course is called Xai, is using natural gas
(19:11):
burning turbines to power the AI facilities. We all know
that these AI data centers require a huge amount of
electricity in order to work. So one of the things
that companies that are using these kinds of power sources
that like natural gas burning turbines, one of the things
(19:32):
they're supposed to do is file a whole bunch of
paperwork to get the right permits to actually operate that
kind of machinery, and according to the Southern Environmental Law Center,
they say that there's no record that Xai actually followed
that process in Tennessee. So in an open letter, the
(19:53):
organization claims that Xai has twenty gas turbines operating without
first applying for those permits and has called upon the
local health department to investigate Xai to determine if, in fact,
that is what is going on now. I can't say
I would be surprised to hear that a Musk backed
company jumped into a situation like that without proper paperwork,
(20:15):
but I have to be fair. That is my bias talking,
Like just because Musk has proven that, you know, he's
done that kind of thing before, it doesn't mean that
that's what's happened this time. It could all be totally
a misunderstanding, like maybe the paperwork has been filed and
they just didn't find record of it. But they are
calling for an investigation to find out if the company
(20:37):
in fact followed the right process in order to operate
this way, because the concern here is that the operation
of these twenty gas turbines could contribute to pollution in
the Memphis area. Meanwhile, the Center for Countering Digital Hate
released the results of a study focusing on the GROCK
AI assistant that's the AI assistant connected to YEP, formerly
(21:00):
known as Twitter. YadA YadA, YadA so. This study concluded
that GROC lacks sufficient guardrails to prevent people from using
it to generate misleading images of real public personas, including politicians.
The study said that the tool was able to create
a convincing fake image of Donald Trump appearing to be
very sick in a hospital bed, and another of Kamala
(21:24):
Harris appearing to do drugs. The researcher said they tried
various methods to generate these images that range from just
blatantly giving a text prompt asking for what they wanted
to see, to taking a more circumspect route where they
used descriptive language to get the same effect, to get
the same results they wanted without actually triggering any preventative
(21:47):
safety features that otherwise would stop that kind of request
from working. I find this particularly frustrating, not just because
of the disinformation aspect, which is clearly bad enough alone,
but also because Elon Musk like again, I don't like
the guy, but he has repeatedly sounded the alarm about
(22:08):
the dangers of AI, and I agree AI is potentially
very dangerous if it's implemented poorly. And he has argued
that AI needs to be developed and deployed responsibly. I
agree with that too. I think that AI does need
to be developed and deployed responsibly, and that there should
be plentiful safety measures in place to prevent AI from
causing harm. I agree with that. All of these things
(22:31):
I agree with, And yet here is an AI tool
on one of Elon Musk's own platforms that clearly fails
to follow any of those directives. But then you could
also make the same argument that the AI found in
Tesla vehicles has at least appeared to be less reliable
than the company tends to market to potential customers. So
(22:54):
there's that as well. There's also been a recent change
to Grock, the AI assistant on it, apparently in response
to a complaint letter sent to the company from the
offices of five different Secretaries of State, so those states
would be Minnesota, New Mexico, Washington, Pennsylvania, and Michigan. The
(23:14):
offices were complaining that Groc was giving incorrect information in
response to prompts asking about ballot deadlines in those states.
So essentially this is related to election information, and GROC
apparently was just giving incorrect responses when people were asking
these basic questions about ballots. So now rather than just
(23:37):
giving an incorrect answer, GROC will respond to those kinds
of requests by urging the visitor to go to vote
dot gov for the actual answer, So that I think
is a market improvement. Tom David S. Pumpkins Hanks has
a message for all of y'all out there, and that
(23:57):
message is to not trust drug advertisements that seem to
have Tom Hanks voicing the ad copy. So specifically, on Instagram,
Hanks posted quote, there are multiple ads over the internet
falsely using my name, likeness, and voice promoting miracle cures
and wonder drugs. These ads have been created without my consent,
(24:21):
fraudulently and through AI. I have nothing to do with
these posts, or the productions and treatments, or the spokespeople
touting these cures. I have Type two diabetes and I
only work with my board certified doctor regarding my treatment.
Do not be fooled, Do not be swindled, Do not
(24:42):
lose your hard earned money. End quote. Now, if all
this sounds a little familiar, that might be because of
something that happened around the same time last year, where
there was this ad video going around with an AI
generated deep fake version of Tom Hanks prom voting a
dental plan Lisa needs braces dental plan. Sorry, sorry, sorry.
(25:06):
Some episodes of The Simpsons are kind of hardwired into
my brain and I just go into like a feedback loop.
I have to admit that when I was thinking about
how deep fakes were going to spread beyond the creation
of pornographic images and video, which is its own thing
and it's horrifying and terrible, when I was thinking about
how it was going to branch out beyond that, I
(25:27):
had not clocked Tom Hanks as being one of the
first high profile victims. But here we are. Bloomberg reported
that Apple is cutting one hundred positions in its digital
services department, and that the affected employees will have two
months to find another position within the company or else
be struck from the roster entirely. So apparently the main
(25:49):
group affected is the books team, which, as Samuel Aksin
of Ours Technica has pointed out, has no subscription service
attached to that particular product. So it sounds like Apple
really wants to redirect energy towards services that more directly
impact revenue. Which I mean, you can understand that from
(26:10):
a business perspective, right, Like, why would you focus on
things that don't impact revenue? Maybe they impact the value
or the perception of value, which you could argue is
a reasonable reason to keep it going. But yeah, most
companies are going to say revenue is way more important.
(26:32):
So it appears that Apple is one of those. I
hope anyone who has been affected by this will find
a new position with a great team as soon as
is possible. It's been a pretty long time since I've
talked about Uber getting hit with fines, but this next
story is quite a doozy. In fact, according to ngadget's
(26:52):
Sarah Fielding, it would be the largest fine Uber has
ever faced, a whopping two hundred and ninety million Euro penalty.
That's the equivalent of around three hundred and twenty four
million dollars. So what the heck happened? Well, this has
to do with the EU and GDPR. GDPR that's the
(27:13):
set of rules that are meant to protect EU citizens'
personal data. So one thing that companies are not supposed
to do under these rules is to send the personal
information belonging to EU citizens outside the EU itself unless
you're using appropriate measures to make sure you're protecting their information.
(27:34):
This is a thing that a lot of platforms have
run into where they were found to be transmitting information
to servers in the United States, but the information belonged
to EU citizens, and without those protections in place, these
companies can't guarantee the safety of that information. Because GDPR
is so focused on the safety of personal information, that's
(27:57):
a problem. So the Dutch Data Protection Authority or Dutch
DPA says that Uber really failed to protect information for
Uber drivers and that included stuff like their taxi licenses,
their identity documents and lots more other information. That Uber
transmitted this info overseas to Uber data servers in the
(28:20):
United States while not using any approved transferred tools that
are meant to ensure the safety of those data transfers.
The Dutch DPA said this failure to provide adequate protection
is a violation and that Uber subsequently has to pay up.
And it's notable here that as far as I can tell,
no unauthorized individuals ever got access to this information. This
(28:43):
isn't like a data breach or something, so it's not that,
you know, Uber failed to protect the info and someone
got hold of it. It's that Uber was not following
the approved process to make sure data was handled appropriately,
so arguably that personal information could have been intercepted or
exploited by other parties. And Uber reps have said that
(29:04):
the company will of course appeal the fine, so we'll
have to see where this develops from here. Okay, I've
got some more news stories to talk about. Before we
get to those, let's take another quick break. Okay, let's
(29:25):
wrap up some of our tech news this week over
on blue Sky, which, in case you're not familiar with
blue Sky, it's an alternative to Twitter, slash x, along
with other similar tools like Mastodon or threads. So blue
Sky has added some new anti toxicity features and I
think the features are interesting. I'm not sure that it's
(29:49):
the best approach, but they are interesting. It's hard to
determine how to best implement anti toxicity measures on a
platform that allows people to just join up. I mean,
as platforms get larger, they become more difficult to moderate,
assuming that you have a platform that's interested in moderation
in the first place. So yeah, I see also Telegram.
(30:11):
But the big one that I wanted to talk about
was that users are able now to remove their original
post if someone else quotes that post. So in other words,
let's say, let's say that you post something that says like,
I really enjoyed the Barbie movie, which is awesome. Honestly,
(30:31):
I enjoyed the Barbie movie. But let's say that I'm
not me. Let's say that I am the Evil Mirror
Universe version of me, and I've decided, because you have
claimed that you enjoyed the Barbie movie, you are deserving
of criticism and trolling and all that kind of stuff.
So I then quote your post, and I then put
(30:55):
some really nasty commentary underneath about how I think you're
a dumb dumb for liking a perfectly good movie. And
now you can actually take action on this, so before
you might just end up being flooded with like toxic notifications,
let's say that Evil Mirror Universe me has a lot
(31:15):
more notoriety and a lot more followers on Blue Sky.
So I post this negative takedown of your post and
everyone dogpiles on top of you. Now on Blue Sky,
you could go in and remove the quoted tweet from
my post. So you know, if my post is centered
(31:37):
around your earlier post, and you remove your earlier post
from mine, then all it's left is the commentary. If
that commentary is not specific, people don't even know who
I'm referring to as I rant about someone daring to
like the Barbie movie. So that is effective now, granted,
if in my narrative, if in my common Harry I
(32:00):
have also identified you, then removing your post from the
quotation doesn't really help that much because people can still
see who I'm talking about. Maybe they don't get a
link to your profile or whatever, but there's limited utility here.
But yeah, I think that that's generally a positive step.
(32:21):
I'm not sure that's the most effective one, but it
is meant to prevent instances of dog piling. So that
if someone takes your post and then, out of context,
tries to argue that you're a terrible person or whatever,
you can just remove the quotation part out of that post.
Now you can't touch anything else that's in that post, right, Like,
(32:43):
if in my description, I really drag you. You can't
go in and change my description, like you can't change
the commentary I have added. You can just remove the
quoted part, which I think makes sense, Like you shouldn't
have the ability to be able to change what someone's
post reads, because then it's just chaos. But I think
(33:05):
it is a good step. I just think it's tough,
like this is a hard problem to fix. Another similar
feature on blue Sky is that you can actually hide
replies to your posts. So let's say you post something
and no one's quoting you, but they are replying to you,
and some of them are being real nasty. You can
actually hide those replies, and you can choose to hide
(33:25):
it just from yourself, which is like blocking or muting someone,
or you can choose to hide that reply from everybody
so no one can see that this person is just
being nasty towards you. So when someone's being hateful, you
can hide so that no one sees it. Or maybe
you just shield yourself and you let it stand for
everybody else. Blue Sky says it will continue to introduce
(33:48):
features to make the platform a more safe and welcoming place.
This I think is in stark contrast to X like
over at X. You don't really see a Musk and
company putting in measures to make X a safer and
more welcoming place. You do occasionally see stories like the
(34:08):
one I mentioned about NPR, where it appears that whenever
they do enter into content moderation, it's to suppress any
any kind of viewpoint that doesn't go into alignment with
what Musk believes. That at least that's the appearance. Whether
that's what's really going on or not, I don't know.
Like I said, it could really be that there was
(34:30):
just an error in a false positive, But yeah, Blue
Sky strikes me as taking kind of like a maybe
not a total one to eighty approach from X, but
a substantial deviation from that approach. The Motion Picture Association
announced this week that through partnering with authorities in Vietnam,
they have shut down an online media piracy ring that
(34:53):
had accumulated billions of visits between them from January twenty
twenty three up to June twenty twenty four. So sort
of the hub of this, like the central crux was
F Movies, and under F Movies, there were several piracy
sites that were in operation. They included any Wave Flicks
(35:15):
Tours my Flixer and others like that. So the MPa
referred to f Movies as the mothership of piracy and
that the takedown had a cascading effect across these various
piracy sites which have been hosting media illegally for years
and distributing it to whomever goes there. Various media companies
(35:39):
and organizations that represent media companies have spent decades fighting
online piracy, often in ways that end up being outlandishly disproportionate.
Like you might remember, I mean I remember because I
remember the nineties when the RIAA really began to shake
down like service pro like Verizon to hand over the
(36:02):
identities of customers that the RIAA suspected were guilty of piracy.
There are people who were hit with truly ridiculous, astronomic
fines for piracy. And then there's the various issue of
media organizations arguing that piracy equates to loss of revenue,
(36:26):
like if you watch a movie illegally, if you download
a music track illegally, that means we lost revenue. A
lot of entities, including the United States Government Accountability Office,
say that argument doesn't hold water. At best, it is misleading.
At worst, it is wrong, and that's because it's impossible
(36:47):
to determine if someone who pirated a piece of media
otherwise would have purchased that piece of media legally. Let's
say that piracy was impossible somehow, Like we're in a
world where it's just impossible to pirate stuff. Someone who
would have pirated things, you can't say that they will,
though otherwise go out and buy it. Maybe they just
go without. You can't make the argument that a downloaded
(37:10):
piece of media equates to loss of revenue, So a
lost sale isn't the same thing as stolen money. Like
if I walk up to a movie theater and I
look at the list of films that are available, and
I ultimately decide, eh, I don't feel like watching any
of these, and I walk off. I didn't steal from
(37:30):
any of the movie companies. So yeah, I think piracy
is bad. Just to be clear, I do think piracy
is bad. I do not think you should do it.
Don't be a pirate. However, I think media companies that
take the nuclear option based on unsupportable arguments are bad too.
No one's in the right in that particular situation. But
(37:53):
you can read more about this particular story about the
Motion Picture Association and f movies both On Technica and
The Verge. Both sites have some great articles about this
recent Hollywood war against online piracy. I recommend you go
check those out now. For other recommended reading this week,
I've got a couple of different pieces I want to
talk about. First up is a paper that was published
(38:16):
in Nature, the science journal Nature. This paper is by
Valentine Hoffmann, Pratyusha rha Kaluri, Dan Jorowski, and Sharise King.
It is titled AI generates covertly racist decisions about people
based on their dialect. So, as you might suspect, this
(38:37):
paper investigates the issue of bias found within AI models,
but this time with regard to dialect as opposed to say,
physical features. I think it's a really important piece, particularly
since we live in a world in which code switching
is a really important mechanism for navigating different cultural groups
for a lot of people, and if they are switched
(38:57):
into one particular culture, it's entirely possible that an AI
model is going to make some very racist conclusions based
on that anyway. Next up would be a piece by
Heather Knight and Kate Conger of The New York Times
titled Elon Musk's X is leaving San Francisco. City officials
(39:19):
say good riddance. So this story covers how X, which
has long had its HQ in San Francisco, California, is
now moving to other parts of California, like it's going
to rely on offices that are in other cities in California,
but the HQ itself is relocating to Texas, And how
some people who are in various positions in San Francisco
(39:41):
government aren't exactly wringing their hands at x's departure. Then finally,
there's a concerning article in The Texas Observer by Francesca
du Nunzio titled Texas State Police gear up for massive
expansion of surveillance tech. It serves as a reminder that
it's very important to hold law enforcement agencies accountable and
(40:03):
to investigate whether their reliance on certain technologies is merited,
particularly since some of those technologies have proven to have
flaws that can lead to disproportionate harm being done to
vulnerable populations. Specifically, I'm thinking about things like facial recognition
technologies and how that often can disproportionately affect innocent people
(40:23):
from the black community. We've seen that happen numerous times,
So any surveillance state stuff always makes me nervous. It's
also wild that Texas identity is so strange to me
because there's kind of ingrained in Texas is this idea
of independence and being able to stand on your own,
(40:44):
to hike yourself up by your bootstraps, and that you know,
like a lot of the narrative around Texas government is
about being self sufficient. I mean the state of Texas.
Their power grid is independent of the national power grid
in the United States, for example. So you've got this
idea of individualism and self sufficiency, and on the flip side,
(41:07):
you've got this growing surveillance state in law enforcement, and
the two don't really gel together to me, apart from
a slow moved more toward authoritarianism, which I think is
pretty concerning you. But the tech enabling that is really
the feature of that particular article, So check that out.
(41:30):
I hope all of you have a great weekend here
in the United States, the holiday weekend. I hope you
have a great holiday if you're here in the US.
If you're not in the US, I hope you just
have a great weekend period. And I'll talk to you
again really soon. Tech Stuff is an iHeartRadio production. For
(41:53):
more podcasts from iHeartRadio, visit the iHeartRadio app, Apple Podcasts,
or wherever you listen to your favorite show still
TechStuff News
Host
Jonathan Strickland
Popular Podcasts
1. Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
2. Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations.
3. Crime Junkie
If you can never get enough true crime... Congratulations, you’ve found your people.