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, October eleventh, twenty twenty four,
(00:27):
and this week marked a new stage in the ongoing
battle between the United States Department of Justice and Google
slash Alphabet, Alphabet being the parent company of Google. And
this is overcharges of Google engaging in anti competitive practices,
including how Google pays other companies such as Apple to
(00:49):
feature Google Search as the default search tool on devices
like iPhones. So the stage we are now entering into
is the negotiation stage, as the DOJ has released proposed
remedies to address what it sees as illegal activity. Previously
the summer, a judge found that Google had been engaged
(01:10):
in anti competitive practices. So this is the next step, right,
what do we do about this now? On the DOJ's side,
the proposed remedies range from a ban on Google being
able to broker these kinds of deals in the first place,
to going so far as to call for breaking up
the company. Into smaller pieces. Google meanwhile, suggests that the
(01:33):
DOJ should pound zand at least when it comes to
stuff like calling for Google to break up or to
quote Google's Lee and Mulholland quote, this case is about
a set of search distribution contracts. Rather than focus on that,
the government seems to be pursuing a sweeping agenda that
(01:55):
will impact numerous industries and products with significant un intended
consequences for consumers, businesses, and American competitiveness end quote, which
makes it sound like Google's worried that all of that
capitalism is going to collapse if the DOJ gets its way.
Maybe I'm putting too many words in Google's proverbial mouth
(02:17):
at that point, but anyway, a Casey Newton newsletter, and
as I've always said, I have no connection to Casey,
don't know him personally or anything, but his newsletter is fantastic.
In his newsletter this week, he indicates that this is
really the starting point that these two parties have when
it comes to negotiating an actual remediation strategy with Google.
(02:39):
You know, they got some time to do this. A
judge isn't going to hear the actual case about proposed
remedies until next April, and there won't be a ruling
on the matter until next August. So you could see
this as both sides taking one of those positions where
they're asking for way more than they think they're going
to end up getting, and they know that they're going
(03:02):
to have to meet somewhere in between these two extremes. Plus,
we actually do still have a presidential election as well
as elections for the House and the Senate happening between
now in April, and those could clearly have a big
impact on how the government proceeds. I think we may
see a change in how Google positions itself for the
(03:23):
search market, but I doubt it's going to go as
far as forcing Google to break up. We have seen
cases where the government has come in and forced companies
to break up. Those don't always take There was a
famous one with Microsoft that ultimately got reversed, but I
just don't see that happening here. Maybe it will, and
maybe by this time next year we'll all be talking
(03:44):
about how Google is going to get split asunder into
different companies. I have my doubts about that, but we'll
have to wait and see. Tesla held its long delayed
cyber cab event this past week. It was called the
We Robot Event. WE as inwa E, not WE Robot.
Musk had previously teased this event earlier in the year,
(04:06):
initially planning it for August, but that clearly didn't happen.
As August approached, the date slipped and we got October instead,
and the company showed off a lot of stuff at
the Wei Robot event. Part of that was a Tesla
transport vehicle dubbed the robo Van, but apparently it's pronounced
(04:27):
reboven based upon how Musk said it. Anyway, according to Musk,
the autonomous vehicle will be able to haul cargo or
carry up to twenty passengers to solve for high density,
which I mean, that's kind of what the hyper loop
was supposed to do, especially like what the hyper loop
in Las Vegas is supposed to do. These days, we
(04:48):
just have Tesla's chauffeurs driving people around a fairly limited
area of Las Vegas through underground tunnels, which I think
isn't exactly the hyper loop that all of us were
anticipating when Musk first mentioned it, But what do I know,
Maybe that'll turn out to be actually a great idea
in the long run, I've seen a lot of skeptical
(05:09):
reports about how it doesn't actually solve for any meaningful problems.
As for the cyber cab itself, tech crunches Rebecca Bellen
described it as a quote smaller, sleeker, two seater cyber
truck end quote. Musk claimed that the operation of the
cybercab would be such as to bring the cost per
(05:30):
mile traveled to be at around twenty cents according to
US currency plus. He said the average person will be
able to buy one of these things, with the projected
vehicle cost coming in at under thirty thousand dollars. I
think the vision is to have a citizen owned fleet
(05:50):
of cybercabs, so you could essentially rent out your cab
like it was an uber, but you don't have to
drive the thing, so you're just passive generating income by
using your car as a transportation vehicle for strangers. I
don't know if that would actually work. Maybe it would,
but I suspect you would need to do a lot
(06:11):
of maintenance and cleaning and that that would start to
eat into whatever profits you made. But what do I know.
Musk also claimed the company would be engaged in quote
unsupervised FSD in Texas and California next year. In the quote,
that's really ambitious. You know, unsupervised FSD meaning fully autonomous
(06:31):
driving where you don't have a safety driver there, or
if you do, they're not actively trying to solve for
the car's problems. I mean, unsupervised FSD does suggest a
truly autonomous vehicle. I think that is pretty ambitious, particularly
since Tesla vehicles and Tesla owners, let's be fair, have
(06:54):
garnered a reputation for getting into dangerous situations while in
autopilot or full self drive modes. We've all seen the
stories of various accidents that have happened while Tesla vehicles
in these modes have failed to recognize a dangerous situation
and then gotten into really bad accidents, sometimes with tragic results.
So it does seem pretty ambitious to me to suggest
(07:19):
that next year there's going to be these unsupervised FSD sessions.
But I don't know. Maybe they'll be under very controlled circumstances,
which I could sort of understand, but when they'll I
last covered all this earlier this year the state of California,
the regulators there indicated they had not received paperwork from
Tesla that would be necessary to conduct those kinds of
(07:40):
operations within the state. Now that may have changed since then.
I couldn't find anything about it when I did some cursorysearching,
but I admit I didn't spend a whole lot of
time looking for it. So I will do that because
I'm curious to see if perhaps Tesla actually has secured
the permissions necessary to be able to conduct these kinds
of operations. And I'm sure a lot of other media
(08:02):
outlets will follow up on that story as well, so
perhaps I'll have more to say about this next week.
Musk says he expects the Cybercab to go into production
before twenty twenty seven. Now, on top of the cabs
and the Van, Musk also showed off the company's humanoid robot,
dubbed Optimus. According to one article I read, robots at
(08:23):
this event were serving as bartenders as well as dancers,
and according to Bellin's piece and tech Crunch, the projected
cost of one of these things is somewhere between twenty
and thirty thousand dollars, so it's not exactly something the
average person is going to be able to splurge on
I imagine, But then keep in mind Elon Musk is
frequently in the running for richest person in the world,
(08:45):
so to him twenty or thirty thousand dollars isn't the
same as it would be for I don't know anybody else. Now,
the event has only happened since, you know, yesterday it
was it was on Thursday when the event was held. However,
there are already pieces about how investors are disappointed in
the event. Most of the reports I saw said that
investors felt like it didn't focus enough on the business
(09:08):
side of things and instead came across as just sort
of a marketing event. And as I said, Musk already
has a reputation for talking a big game but then
failing to follow through. I mean, he's been talking about
fully autonomous vehicles being a thing since twenty sixteen at least,
and it's always just been like a year or two
(09:28):
in the future, and those years come and go and
we're still not there yet, So maybe that's contributing to this.
In pre market trading, the share price fell around six
percent for Tesla, but that could just be a momentary blip.
In fact, by the time you listen to this episode.
It could be that the market has completely changed around.
I don't know, but you know, he could take a
(09:49):
quick surgency. Okay, we're going to take a quick break.
When we come back, I've got more tech news to cover.
We're back. I got a couple of AI related stories
to talk about this week. First up, Hayden Field of
(10:10):
CNBC reported on how OpenAI warns that quote bad actors
are using its platform to disrupt elections, but with little
viral engagement end quotes. So, in other words, the fear
that people would turn to AI in order to manufacture
misinformation appears to be well founded. That is what is happening.
But apparently, while there are people who are using AI
(10:32):
to generate misinformation, that misinformation isn't getting very much traction.
So in a pretty long report, open ai said it
has detected and intervened in more than twenty misinformation campaigns
that were using open AI's products like chat GPT. These
campaigns included stuff like, you know, social media posts that
(10:53):
were made by bots that were asked to create ultimately
malicious posts to AIG generated web content designed to push
specific narratives in an effort to shape perception and support,
and it sounds as though the use of this AI
for this kind of purpose it's on the rise, and
that's no big surprise, especially now that the United States
(11:14):
is about to hold an election. There's a lot of
stake in the game if open ai is to be believed. However,
the good news that we can take away from all
this is that the AI generated stuff isn't getting any
you know, viral attention. Now, if that is true, that's great,
but we have to remember that the whole goal of
generative AI projects is to get to a point where
(11:36):
the stuff that AI is generating is as good, if
not better, than the stuff that humans make. So at best,
you could say this is really more of a slam
against open AI's tools, saying they're not good enough to
create viral posts like that's a message that I think
open ai needs to be aware of because if they're saying, hey, yeah,
(11:56):
sure people are using this to create misinformation, but here's
the good news, it's not really effective, then you could say, well,
does that mean that your tools can't make effective persuasive messaging?
Because if that's the case, why would I use it
in order to I don't know, design a marketing campaign
for my product. That's tricky, right, because if you say no, no, no,
it's good for that, then okay, well why is it
(12:17):
not good at creating misinformation? Or are you just trying
to be all things to all people while avoiding as
much responsibility as possible. I don't know the answers to
any of that. I just think it's interesting. Meanwhile, The
Guardian's Ben Macouch reports that Russian campaigns are using AI
generated images to spread disinformation about hurricanes Helene and Milton,
(12:39):
which just recently impacted the southeastern United States. For example,
an account on x the app formerly known as Twitter,
posted images of a supposed flooded Magic Kingdom at Walt
Disney World in Florida. Now, any real Disney Parks fan
knows the Magic Kingdom is actually built on the second story, or,
(12:59):
if you're from Europe, the first floor, because there's actually
a series of corridors known as the Utila Doors that
are underneath areas like Main Street USA. So if you
had really significant flooding, I'm talking like the stuff so
deep that you're having to wade through it, that would
be a heck of a thing. In the Magic Kingdom,
it would probably mean those utilid doors would have to
be almost totally flooded, which would be absolutely disastrous. Anyway,
(13:22):
the images were actually created by AI, which many on
x pointed out, and ex itself relatively quickly acknowledged it
by tagging the posts saying that the images were AI generated.
And there's been a lot of misinformation surrounding the hurricane
season in the United States this year, and it appears
that Russia is trying to stir the pot further with
(13:44):
attempts like these, which you could argue are fairly feeble,
like they're not really effective. But I think the philosophy
of the Kremlin is if there's already fire, pour some
gasoline on it. As we get closer to a deadline
imposed by the United States government on Byteedance, the parent
company of TikTok, and that deadline is that byte Dance
(14:05):
must divest itself of TikTok in January or face a
nationwide ban on the app here in the United States.
ByteDance has reportedly started to cut staff by the hundreds. Now,
this plan, according to Reuters, is to lean more heavily
on artificial intelligence for the purposes of moderating content Reuter's
(14:25):
reports that according to TikTok, around five hundred positions were
cut in Malaysia. Other sources told Reuters that was closer
to seven hundred positions. But yeah, another case of AI
cutting into the jobs of human beings, in this case
content moderation, which is already a pretty sensitive topic, right
because there's a lot of concern that platforms in general,
(14:47):
not just TikTok, but kind of across the web are
not doing a good enough job when it comes to
moderating content, doing things like protecting from harmful content to misinformation.
TikTok is also in the new here in the United
States because, after an investigation spanning more than two years,
legal documents that were supposed to be sealed that are
connected to a lawsuit that alleged TikTok has engaged in
(15:10):
practices that put young people at risk in various ways.
Those documents now reveal the company had its own internal
investigations that seemed to support those allegations. That is, TikTok
had conducted research on issues relating to child safety and
found that indeed the company's practices were likely causing unhealthy
consequences for younger users and you know, they ended up
(15:33):
doing that stuff anyway. So this is kind of like
the whistleblower situation over at Facebook now known as Meta,
in which internal documents showed that at least some people
within the company were really concerned about how products could
affect users, but the company at large seemed to dismiss
those concerns, and PR reports that the details in these
(15:53):
documents were intended to be redacted, but that an error
in publication meant that large portion of the documents were
not redacted, and before the court could remedy that, the
Kentucky Public Radio published some of that material. That's where
it was revealed that executives within TikTok were aware of
(16:13):
potentially harmful practices and how some remediation strategies were really
nothing more than lip service, and yet the company promoted
these features as being a meaningful way to address concerns
like time management type features, right, the idea of let's
try and alert people so that they limit their time
on the app. According to the documents, it shows that
(16:35):
people within TikTok realized this wasn't actually discouraging people from
using the app as much. They still were using the
app a lot, but it did give TikTok a little
bit of smoke screen to make it look like the
company was doing something about an issue that was concerning
a lot of folks. I actually really recommend reading nbr's
full piece on this story. They go into a lot
(16:56):
more detail. The story is titled States probed TikTok for Years.
Here are the documents the app tried to keep secret,
and you can find that on NPR dot org. It
originally published earlier this morning, Friday, October eleventh, twenty twenty four.
Chinese based hackers are said to have successfully penetrated networks
(17:16):
belonging to US Internet service providers, including Verizon and AT
and T. And they did it by targeting backdoor surveillance
systems that are meant to allow US agencies, you know,
like the FBI or the NSA a way to conduct
surveillance across these networks. And it's almost like an entire
population of people, including myself, have pointed out repeatedly that
(17:39):
intentionally creating workarounds to secure systems means you've turned them
into inherently insecure systems. Right. I use an analogy. I say,
imagine you've got a bank vault and you've got the
world's most sophisticated, powerful vault door that cannot be breached.
It is incredibly secure. But you've also put a screen
(18:00):
door in the back of the vault so that the
FEDS can sometimes just pop in and out just to
make sure everything's okay. Well, if the thieves know that
there's a back door, you know they're gonna dedicate their
time to finding the back door and using that rather
than go through the front, because going through the front
would require way too much time and resources. Just find
the way through the back. It's already there, you can
(18:22):
just use that. That seems to be what happened. So
Chinese hackers being called Salt Typhoon have apparently done this
and were able to access the various ISPs potentially for
months before they were detected and they were able to
be addressed. So yeah, that's a real issue. As to
(18:42):
how bad that hack was hard to say at the moment.
There's still investigations that are ongoing, but it's not like
this was some crazy unforeseen outcome. People have been warning
for years that these intentional back doors are a bad idea. Okay,
the one bit of news I missed last week that
I wanted to mention just to wrap this up, is
(19:04):
that Microsoft last week announced it had discontinued the HoloLens
augmented reality headset. It's no longer producing new units, and
it plans to in support for existing devices by December
twenty twenty seven. So the hollow Lens launched in twenty sixteen,
there have been some preview events the year before, and
it was never a consumer product. Microsoft marketed it more
(19:25):
as an enterprise product for businesses. The company released an
updated model in twenty nineteen. The headsets cost around three
thousand to three five hundred dollars, so they weren't cheap,
and it looks like Meta is going to be the
one charging ahead with its own AR headset development, while
Microsoft and potentially Apple take a step back from attempting
(19:46):
to augment reality. That's it for the news this week.
I hope all of you out there are doing well,
and I will talk to you again really soon. Tech
Stuff is an iHeartRadio production. For more podcasts from iHeartRadio,
(20:06):
visit the iHeartRadio app, Apple Podcasts, or wherever you listen
to your favorite shows.