November 1, 2024 • 43 mins
We knew that the 2024 election season in the US would include some tough stories involving tech and potential election interference. In this episode, we look at a few headlines of how tech may have already played a role in misinforming voters either on purpose or by accident. Plus many more headlines in this jumbo news episode!
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, John van Strickland.
I'm an executive producer with iHeartRadio. And how the tech
are you. It is time for the tech news for
the week ending on November one, twenty twenty four. And
(00:26):
here in the United States, we are rapidly approaching election day,
which is Tuesday, November fifth. Now, for those of y'all
eligible to vote in the United States but who have
not yet done so, I urge you to get out
there and vote. Do your research, look into candidates and
the issues in your area so that you can make
an informed decision, and go deeper than just sites that
(00:48):
are run by politically aligned outlets. Look for information that's
as objective as possible. It's very easy to fall into
a trap of misinformation or bias, and that goes for
any side. I'm not picking specific sides here, it's true
across the board, but for goodness sakes, get out and vote.
It's one of the most important components in maintaining a democracy,
(01:11):
and the very act of voting is under assault these days.
And I don't know about you, but I would rather
continue to have a say, even if it's just a
tiny say in how the country I live in operates.
Now that we have that out of the way, let's
get to a whole bunch of tech stories that relate
to the election, because there are tons of those this week,
which is no big surprise. First up, some voters in
(01:34):
Pennsylvania received a strange text message this past weekend stating
that they had voted in the November fifth election, even
though they hadn't done that yet. Now, obviously that could
stir up some pretty intense confusion and negative emotions. If
you received a message saying that you had already voted
(01:55):
and you had not already voted, clearly you would suspect
hanky and panky going on. Now, in this case, it
wasn't any sort of hanky panky at the actual polling locations.
It wasn't like the state department messed up. Instead, it
was with an organization called All Vote, and All Vote
had previously gotten into hot water for texting registered voters
(02:18):
to alert them that they were not actually registered to vote,
when in fact they were. County officials in different regions
of Pennsylvania have warned that that particular approach was a
hoax designed to gather personal information from people that would
later be exploited against those people. Max Marvin and Kate
Bernard reported on this issue in the Philadelphia Inquirer in
(02:41):
a piece titled Pennsylvania residents got phony texts claiming they
had voted already Ignore them, officials say so. They also
quoted a contracted spokesperson for All Vote named Charlotte Climber.
Climber has claimed that this is all just an innocent
but incredibly dumb and careless mistake made when All Vote
(03:03):
employees entered some sort of typo into their system. So
apparently the original idea was that the organization would text
folks who had voted in the twenty twenty two election
and direct them to voting resources for the twenty twenty
four election. Now, why All Vote felt the need to
inform people that the vote they had cast two years
(03:27):
ago had in fact been counted is a bit beyond me.
That ship has sailed Like I guess the message would be,
you voted in the election, meaning the twenty twenty two election.
Here are some resources for you. But because it left
off the twenty twenty two part, it seemed to indicate
that people had already voted in this election. Now, that
seems to me like that messaging was unnecessary in the
(03:50):
first place. And if that messaging had been left out,
you know, the whole you had voted in the election bit,
there'd be nothing to talk about with this story. It
would just be a text message saying, hey, here's some
resource for you for voting in this election, and then
there'd be nothing to talk about. Climber has said the
founders of All Vote are progressives, but declined to give
more information about them, saying that there's a concern for
(04:12):
their safety and that people who might be hard right
leaning voters might do them harm. This is where I
say that while I consider myself a progressive, so I
would consider myself aligned with the people who are at
least alleged to be behind All Vote. I also think
All Votes operations are not great like they're not done properly.
(04:39):
They were poorly thought out and executed, and if the
owners of all Vote are actually sincere, they're likely to
cause more harm to their cause than to help it
through these kinds of operations. Unless this was an actual
attempt to confuse and discourage people in say hard right counties,
from voting, in which case that's just despicable, Like, I
(05:01):
don't care what side you're for. At least according to
this spokesperson, the people who own all vote believe in
the same things I generally believe in politically, I don't
believe in discouraging people through deception so that they don't vote.
I think that's reprehensible. Whether or not that's what was
going on, I don't know. It could be that this
(05:22):
really was an innocent mistake and just a poorly thought
out campaign in the first place. Either way, a great
way to use technology to discourage the participation of the
democratic process. Speaking of unintentional tech issues affecting the election,
less talk about Colorado. So it seems that the Colorado
Department of State committed a major booboo, or, as Todd
(05:44):
Feathers of Gizmoto called it, a bit of an uh oh.
This comes from an article written by Todd Feathers. It's
titled Colorado agency improperly posted passwords for its election system online. Yeah,
that's not good. So what actually happened? Well, the Department
of State had a link to a spreadsheet on its website,
(06:05):
and the spreadsheet included some hidden tabs, so upon casual glance,
there's nothing hinky going on. However, you can just choose
an option to unhide tabs, in which case one of
those tabs contained passwords for the state's voting machines. Yikes,
I hear you say. Now, it's bad. There's no sugar
(06:27):
coating it. That's bad. That's a terrible breach of security etiquette.
It is, not, however, catastrophic. Why Well, for one thing,
these voting machines actually require two passwords. It's kind of
like one of those things you see in movies where
you have to have two keys to activate the nuclear
arsenal or whatever, and you have to have two people
(06:48):
across the room from each other insert and turned the
keys simultaneously to activate it. Well, in this case, the
voting machines require two separate passwords, and those passwords are
not collected together. So while a set of passwords were
available according to the spreadsheet, the other set was not available.
So even if you had access the spreadsheet and revealed
(07:09):
the tabs and you wrote down all the passwords, you
would still only have one half of the password component.
But that's not at all like in order to actually
use these passwords. You would need to have physical access
to the voting machines themselves. Now, in order to do that,
you would need credentials to be able to get access
to the areas where the voting machines are stored. They're
(07:31):
also under a constant video surveillance, so you'd have to
deal with that as well. Plus, Colorado uses paper ballots,
so there's a physical paper trail of every vote cast.
Voters verify that their ballots are correct before they cast
their vote, so even if someone did access the machine
somehow and messed with them, the paper trail would remain
(07:54):
a resource for verifying the votes that had been cast.
So while the mistake is a giant whoopsie and absolutely
should not have happened, the voting process itself is still reliable.
Of course, that's not stopping some people from using this
to bring into question the results of any votes that
are made in the state. In other words, this is
(08:15):
an opportunity to undermine confidence in the results of the
election ahead of time, so that no matter who wins,
you can have some wiggle room in case your side
didn't win. If your side did win, you say, hey,
look the system worked, and if your side didn't win.
You're like, this is because our system was compromised. Now,
I'm here to remind y'all that for democracy to fail,
(08:36):
you don't actually have to prove that the system is broken.
All you have to do is convince enough people that
the system is broken to reduce their confidence in the system,
and then democracy dies. That's how that works. Reporter Barbara
Ortute has a piece in AP News that's titled report
says crowdsourced fact checks on X failed to a dress
(09:00):
flood of US election misinformation. Now, I think that comes
as a surprise to absolutely no one. Before Elon Musk
carried us SYNC into the HQ of what was then Twitter,
I maintained that was foreshadowing that he would eventually sink
the company, which hasn't happened yet. But goodness gracious, I mean,
I think the band is playing while the Titanic is sinking. Anyway,
(09:23):
the site had content moderators, and they did stuff like
act on tweets that were spreading misinformation, like they could
label tweets as saying this is inaccurate or this is misinformation,
whatever it may be. Those days, however, are pretty much gone.
X relies very heavily on the user community to flag
posts that contain misinformation. You know, these are folks who
(09:45):
are not necessarily trained to do that, or it might
include people who are actually interested in spreading misinformation to
start with. Now, to be clear, this community based tool
existed before Elon Musk took over Twitter. This is not
like Twitter did this in instead of content moderation. Now,
originally this tool was called birdwatch. But while it used
(10:05):
to be a supplemental way for Twitter to monitor and
moderate content, it's now taking a more prominent role because
content moderation is not really a big thing over at
X these days, there aren't that many people left to
actually do it. The Center for Countering Digital Hate took
(10:26):
a look at the community notes tool on X and
found that seventy four percent of the tweets the posts
that they sampled had community notes that were inaccurate or inadequate.
The sample size was two hundred and eighty three posts
that had community notes attached to them. That means that
(10:46):
two hundred and nine of those posts were not accurate.
Now X maintains that the community notes feature is a
useful tool and has pointed to other academic research, saying
that it's a trustworthy source of information. So this is
not to say that definitively the tool is broken or
not performing as intended, but that this one nonprofit group
(11:09):
says that that's the case. For more information and a
more thorough look at all this, I recommend reading an
article by Ashley Bellinger on Ours Technica. It's titled Toxic
X users sabotage community notes that could derail disinfo report says. Okay,
while we're on the subject of Elon Musk, which is
(11:30):
tangential to X, I suppose in PRS Bobby Allen has
a piece titled and Elon Musk backed political group is
posting fake Kamala Harris ads on Facebook. So yeah, we've
got more misinformation and election news headed your way. My
apologies that we knew this was going to happen. I mean,
I've been talking about for months. But in this case,
(11:52):
Elon Musk has provided significant funds to a group called
Building America's Future. This group, in turn, has operated a
Facebook account that is called Progress twenty twenty eight, and
this account has published campaign ads that look like they're
from Kamala Harris's campaign, but in fact they are purposefully misleading,
(12:14):
both to trick the viewer into thinking that the ads
are actually from Kamala Harris's campaign, and also by misrepresenting
Harris's position on various topics. Now, according to the First Amendment,
there's nothing illegal going on here. There's nothing that says
you have to be truthful when you're advertising with political speech. Now, sure,
(12:35):
you got to be truthful when it comes to advertising
products and services. For example, if I were to endorse
something but it turned out I never actually used whatever
it was I was endorsing, I could potentially face a
pretty massive fine for that. Like I could get hit
hard for doing that, My company could get hit hard
for doing that. But you know, when it comes to
(12:56):
informing people about, you know, where politicians stand on specific topics,
truth just isn't a requirement here in the US. Further,
using the name Progress twenty twenty eight for the group
further obvious skates who's actually behind the campaign. So it
would be easy for someone to believe this is a
real ad from the Harris side. And it's interesting because
(13:17):
there are limitations for this kind of thing, I mean,
the use of deep fakes, you know, where you're making
it seem like someone is saying or doing something that
they actually did not do. That could be a violation
of policies. But an ad that ultimately does a similar
thing but without the use of deep fakes, that's fine,
which just makes me curious where the line is. Okay,
(13:38):
we've got tons more information about technology, including within the elections,
but also other stuff. But first, let's take a quick break. Okay,
we're back. Meta held its third quarter earnings call this week,
(14:01):
and CNBC's Jonathan Vanian has a piece all about the
Reality Labs part of that call, which was just one part.
There were there were more things to talk about than
just Reality Labs. But you might remember that Reality Labs
focuses on all things mixed reality, so that includes virtual
reality and augmented reality as well as the metaverse. You know,
(14:23):
this supposed future of our computational approach, and you know,
this has really served as a great way for Meta
to spend tons of money while having to assure investors
that the metaverse is very much going to happen and
it will be the future of computing, and that's something
that stakeholders remain skeptical about. I think for good reason.
(14:45):
That's my personal opinion. I remain unconvinced that the metaverse
is going to be a significant thing. I really doubt
it's going to be the new means of interacting with computing.
I just don't think that's true, but it could be wrong.
So anyway, the call revealed that Reality Labs had a
twenty nine percent increase in revenue, reaching two hundred and
(15:05):
seventy million dollars this past quarter. That's no small shakes.
Two hundred seventy million dollars is a lot of money.
Keep in mind, Reality Labs most of the money they
make is through product like the Meta VR headsets or
the augmented Reality ray bands that Meta has co branded
with ray Ban. Those are the source for revenue in
(15:28):
the most for the most part, for the Reality Labs division,
and two hundred and seventy million dollars that's a lot
of money, but it's not as much as analysts had
been hoping for. They had estimated that Meta would hit
around three hundred and ten million dollars in revenue from
the Reality Labs unit. Further, the division as a whole
chalked up an overall operating loss. So yeah, it made
(15:51):
two hundred and seventy million dollars, but it's spent way
way more than that. How much more? Well, the operating
loss is at four point four billion with a B dollars,
So yeah, you make two hundred and seventy million, but
you spent more than four point four billion. Yikes. Zuckerberg
continues to insist that vr AR and the metaverse are
(16:15):
the future, and he might be correct. Right, just because
it doesn't appeal to me and doesn't really seem to
fit with my concept of what computing really is all about,
that doesn't mean anything at all. I'm just one person,
but I've yet to see a demonstration of this technology
that convinces me that the pros actually outweigh the cons.
(16:36):
I mean, you know, feeling immersed is amazing, don't get
me wrong. Like, if you can experience a really immersive
presentation of some sort, that can be really fun and entertaining.
But wearing a headset for more than, you know, like
half an hour is a pretty hard sell for me,
as well as for a lot of other people. And
you know, people who like me are prone to motion sickness.
(16:59):
If the is not done really, really well, it's a
super hard sell. It doesn't matter how good the experience
is right, like, sometimes our tolerance is pretty low, and
I cannot imagine using this kind of technology to do
work right. Maybe leisure activities, but work is hard to
(17:20):
imagine in this particular context. Now, it could very well
be I'm just suffering under the problems of a very
limited imagination and that this really is the future. I
just remain very skeptical. Jason Kebler of four oh four
Media has a piece that's got a great headline. The
headline is Zuckerberg, the AI slop will continue until morale improves.
(17:44):
That's a great play on a popular joke that's typically
about flogging anyway. As you might suspect, Kebler is not
a huge fan of AI generated content flooding various social platforms,
but he quotes Zuckerberg who said during that earnings call quote,
I think we're going to add a whole new category
(18:05):
of content, which is AI generated or AI summarized content,
or kind of existing content pulled together by AI in
some way. And I think that's going to be just
very exciting for Facebook and Instagram and maybe threads or
other kind of feed experiences over time. End quote. So
that's all direct quote, including like the verbal stumbling now
(18:28):
in full disclosure, I actually don't mind using AI to
summarize something that I have already read in order to
like refresh my memory of the most important points, you know,
kind of like using it like a note taking app.
I don't mind doing that, but I do think it's
still vital to actually read the content first and not
just rely on an AI generated summary, because sometimes those
(18:50):
are not accurate or they misidentify what the most important
points are. I have less than zero interest in having
AI generated content populate my social media feeds as it stands.
I hate that a recommendation algorithm decides which posts I
see and when I see them, because I often miss
(19:10):
out on important, timely stuff because Meta thinks that I
should see that Facebook post later on, maybe sometimes weeks
later after it's been posted. Like I posted a picture
of me having voted. I believe that was on October
seventeenth or so. So anyway, it was not long after Georgia,
my area of Georgia had started doing early voting, and
(19:31):
I was still getting people just seeing that picture for
the first time and liking it this week, more than
a week after I had done it. So that tells
me that the recommendation algorithm is not serving this up
in a timely manner. So I don't want the stuff
that my friends post to be buried even further under
(19:52):
AI generated crap. You know, I'm tired of not being
able to see the things that my friends post. I mean,
there are times where someone appears to disappear from my
life entirely. But it's not that they're gone and they're
not posting or anything. It's that Meta has just decided
that I don't need to see their updates unless I
actually go to their page, and even then I'm not
(20:13):
necessarily going to see it in chronological order. And to
think that even this limited engagement could be limited even
more with AI generated junk is really upsetting. So is
Meta just trying to sever relationships between real life friends. No,
I don't think that's true. I don't think that's what
Meta's goal is. What the company is actually trying to do,
(20:36):
is what it's been doing for years, is trying to
keep you on its various platforms for as long as
possible in order to serve you more ads and to
make more money. And if your friends just aren't enough
to keep you glued to Facebook or you know whatever, Well,
by golly, we'll just have the bot's slave away and
the content minds to create stuff that does keep you there,
(20:59):
which is pretty gross. Oh and while Meta is looking
at a more than four billion dollar operating loss on
the reality Labs division, it's also looking to spend even
more money, largely on the AI side of things. So
according to CNBC, Meta plans to raise the low end
of its capital expenditures guidance for twenty twenty four up
(21:21):
to thirty eight billion dollars, So that's up a full
billion dollars from what it was before. Earlier, the low
end for expenditures for this year was going to be
thirty seven billion. Now the top end is forty billion,
which remains unchanged from earlier in the year. So really
the margin has just decreased, right, It's going to be
somewhere between thirty eight and forty billion dollars, but it's
(21:43):
definitely not thirty seven that's too little. Zuckerberg says next
year will be even more dramatic, which kind of is
in line with what he's been saying for the last
couple of years. And as I have said repeatedly on
tech stuff, AI is expensive and running AI operations in
the cloud costs a whole whole lot of money. You
need a lot of data centers, you need specialized processors.
(22:05):
This racks up the cost of operation very quickly. It's
one of the reasons that I actually favor having AI
processing capabilities built into chips that are in consumer devices
and moving at least some or hopefully all of that
AI processing to the end device, because one that speeds
things up considerably because it's just handling your calculations, not
(22:28):
everybody's two. It vastly improves security and privacy because you're
not constantly beaming your information to some server farms somewhere.
As you engage with an AI application, it's not being
held in some server farm. It's not being used to
train the next generation of large language models. It's just
you interacting with your device. So I much prefer that
(22:52):
approach to AI interactions than the cloud based ones. But
that's really a topic for another time now. Like reality labs,
the AI costs are something Zuckerberg doesn't really shy away from.
He just wades right into it. He says it's a
necessary cost in order to establish the infrastructure for future platforms,
(23:12):
and again he might be right, like, maybe this is
the future and he's just getting the work done early,
and it's very painful because it's very expensive and it's
not paying off right away, but that in the long
run it sets the groundwork that allows Meta to prosper.
But in the meantime, the market has been a little critical.
Shortly after the earnings call, in after market trading, the
(23:35):
stock price for Meta took a bit of a dip.
So it seems to me that investors are getting a
little anxious and tired of these reports about the company
spending so much money on reality labs and AI when
the payoff for those investments is pretty far down the road. Personally,
(23:57):
I think in general that preparing for the long term
is a better approach than just focusing on the short term. So,
you know, I feel conflicted about this because I do
think that if you're going to really invest in a
company's future, being willing to do that and know that
you're not going to see returns right away takes a
(24:17):
lot of bold decision making and ultimately can be the
best thing for the company. On the flip side, I
remain unconvinced that the metaverse and AI is really going
to give Meta the ability to stand apart from other
competitors in the space, or that the company has established
that there's a future in that particular type of computing
in the first place. So double edged sword kind of thing.
(24:40):
But I would actually like to see more companies take
a firm stance on yes, this is a painful decision
in the sense that you know, we're investing money in
something that's not going to be an immediate payoff, but
down the road it's going to be the best thing
for our company. I am really tired of seeing companies
that almost sacrifice everything for the purpose of getting short
(25:02):
term gains, because ultimately, I think that ends up corrupting
companies from the inside, and they ultimately crumble under their
own weight once they reach a point where you just
can't do that anymore. Okay, enough about that, But one
other thing that Zuckerbird brought up during that earnings call
was that Threads, the Meta operated alternative to Twitter, is
(25:23):
up to nearly two hundred and seventy five million monthly visitors.
That is a one hundred and seventy five percent increase
from last year, and he also said the app is
seeing more than one million new users sign up every
single day. Now, that, along with a general migration away
from X formerly known as Twitter, means the gap between
(25:46):
the two services is beginning to narrow. Analysts from Sensor
Tower estimate that X has somewhere in the neighborhood of
three hundred and eighteen million users. Now, I say estimate
because X is a private company, and as a private company,
they don't have to share that kind of information. Threads
is somewhere around two hundred and seventy five million monthly users.
(26:07):
So the growth of Threads, along with folks who are
ditching X, has changed the landscape quite a bit. Meta's
CFO Chief financial Officer Susan Lee has said that the
company doesn't expect Threads to be a significant revenue driver
in twenty twenty five. Rather, the focus will be on
growing the platform further before really exploiting the heck out
(26:28):
of it. Now, if I were a betting man, I
would say that Meta will eventually sprinkle some of its
targeted advertising ferry dust on Threads, and it will probably
happen in the not too distant future, and it will
increase Meta's already significant hold in the online advertising space. Meanwhile,
I expect X's decline to continue simply because Elon Musk
(26:51):
cannot help himself and consistently makes decisions that drive away
many users of X to other platforms include Threads and
blue Sky and mast it on. I think the one
advantage X has is that no other singular microblogging platform
has the dominance that Twitter once had, So either users
(27:14):
are adopting multiple platforms and kind of spreading their time
across them, or they're having to choose one of several
different offerings and just be satisfied with the fact that
they're not going to see everything they used to see
because other people have gone to competing services. So that's
kind of my view of where things are right now. Okay,
we're going to take another quick break, but I still
(27:35):
have plenty of tech news. This week was shock full, y'all,
so let's take a quick break. We'll be back with
some more headlines. Kylie Robson of The Verge has a
piece titled open AI's search engine is now live in
(27:57):
Chat GPT, which I'm sure is for concern for some
folks over at Google. Now. To be clear, this feature
is currently only available to paid subscribers of Chat GPT.
It will later roll out to other users, like, you know,
enterprise users or people who are working in educational institutions.
(28:17):
As Robison explains, the feature is integrated directly into the
chat gpt experience, so it's not like a separate tool
or tab or something. Users when they're putting in a
query and conversing with the chat gpt bot will sometimes
receive suggestions for various websites that relate to whatever the
(28:41):
topic of discussion is, so they'll get suggestions for resources
that they can use to further answer their questions. Robison
points out that chat GPT's search results do not contain
promoted results, so that might be a big advantage over
folks who previously had used Google. You can, and also,
by the way, directly ask chat gpt to give you
(29:04):
web results, so you can ask it to give you
those those websites from the beginning and just treat it
like it's a search engine. And by not including promoted results,
that's a shot across the bow at Google, because that's
Google's bread and butter. I'm guessing anyone who has used
Google for like the last decade has noticed that sponsored
(29:25):
results appear at the top of search results. Lists now.
In the very old days, the general expectation was that
the most relevant, highest quality search results would appear at
the very top of Google's results page. But for the
last several years, the very top of that list is
frequently dominated with paid ads that lock in those spots,
(29:48):
And even though Google does tag those search results as
being promoted search results, some folks still bristle at how
this feels like preferential treatment for what is often thought
of as a scholarly like ibbe. As if you went
to the library and you wanted to look up books
that were about a specific topic, and you found out
that the results of all the books in the library
(30:10):
that do relate to that topic have been ordered not
by author name or whatever, but by way of who
has paid the library to have their book promoted the most.
That's not cool. Open ai claims it does not intend
on selling advertising the way Google does, but that does
raise questions as to how AI will generate revenue from
(30:33):
free users who use this tool, because presumably free users
will be able to interact with this in the not
too distant future. Another question that remains is how open
ai will navigate the choppy waters when it comes to
referencing or pulling from various resources online. Most media companies
would very much prefer if you were to go to
(30:53):
their website and read stuff on their own web pages.
That's one of the many reasons I actually recommend folks
check out the articles that I reference in these episodes. BENJ.
Edwards of Ours Technica, as an article titled Google CEO says,
over twenty five percent of new Google code is generated
by AI. That's really incredible that more than one quarter
(31:15):
of the code generated within Google is coming from AI tools. Now,
according to Google executives, all these operations are still ultimately
overseen by actual human beings. AI might generate some of
that code, but humans review and then approve that code
before it can be deployed. Now, the purpose of this
structure is to streamline coding and speed things up, presumably
(31:38):
while also cutting down on the need for quite so
many people coding for Google in the first place. And
as Edwards explains, not everyone is super happy about how
AI is taking such a prominent role in coding. There
are critics who worry that relying so heavily on AI
means we could end up with apps and code that
has more errors in it more vulnerabilities interduced into the
(32:00):
code that are difficult to detect, meaning that anyone deploying
that code could be creating opportunities for hackers, and maybe
the hackers are using their own AI tools to identify
and then exploit those vulnerabilities. In fact, I think that's
just a guarantee. I think AI coding and AI hacking
that's the reality we face now. It's going to be
(32:21):
a seesaw like relationship moving forward. I recommend reading BENJ.
Edwards article in full for a deeper look into the story,
including how some studies have shown that folks who have
used AI to help code will simultaneously end up believing
that their code is superior to other people's code because
they used AI tools to make it, while also introducing
(32:43):
more bugs into their own work. So it's kind of
like creating this false sense of superiority or our expectation.
Jonathan M. Gitlin, also of ours Tetnica, has another Google piece.
This one is titled Generative AI is Coming to Google Maps,
Google Earth Ways and Yeah. It's all about how Google
(33:04):
is incorporating more AI tools into its various map apps
to give users more features. Some of this was news
to me, In fact, quite a bit of news to me,
even though apparently some of these features have been around
for a while. So for example, Getland quotes Google's VP
and general manager of GEO. That's Chris Phillips, who said, quote,
(33:26):
think of features like lens and maps. When you're on
a street corner, you can lift up your phone and look,
and through your camera view you can actually see. We
laid places on top of your view so you can
see a business, Is it open, what are the ratings
for it? Is it busy? You can even see businesses
that are out of your line of sight. End quote.
(33:47):
I didn't know about that, although I have talked about
that very use case and augmented reality in the past.
But I think it's pretty cool, particularly if you're out
and about somewhere that's new to you and you want
to find a place that's like open and nottally slammed.
I actually could have used that. Last weekend, me and
my party were in Pennsylvania for the Pennsylvania Renaissance Fair,
(34:07):
which now is over, but if you ever get a chance,
it's a phenomenal renaissance fair. I say that as someone
who used to work at a Renaissance Fair. It's amazing.
But after the fair was done on Sunday, we wanted
to go and eat at a restaurant, and we had
identified which one we wanted to go to, but by
the time we got there, we found out the kitchen
had already closed, and so we were forced to change
(34:28):
our plans and find somewhere else to grab dinner. Having
this tool would have told us like, oh, by the
time you get there, it's going to be too late,
so don't even bother, and that would have saved us
some trouble. Anyway, Google is bringing AI enabled search results
to Maps, so if you just have a general hanker
in for something but you don't know any more details,
you can ask Maps and have a little conversation with
(34:50):
an AI bought to whittle down your options. Check out
Getland's article to learn more about the upcoming features headed
to Google Maps and these other tools. Joshua Tyler of
Giant Freaking Robot has an article titled I attended Google's
Creator Conversation event and it turned into a funeral. Yikes.
All right, so this is about an event that Google
(35:11):
held this week, and Tyler's description of it is pretty
darn grim. He talks about receiving a tour of the
Google campus, and it was a tour of a campus
that was almost completely empty, although the event was happening
in the middle of a work day. He describes getting
not one but two security badges before being allowed into
a secure building, only to have no one actually look
(35:33):
at or check his credentials once that time came. According
to Tyler, the attendees of this event consisted of folks
who worked on independent websites that have been shadow band
due to Google's changes in its approach to generating search results. Now,
in case you're not familiar with the term shadow band,
it means that you're blocking someone or something from your
(35:56):
online service like a forum or in this case, Google
search results, and they don't have any knowledge that they've
been blocked. So the entity is still allowed to make
posts on the platform. Like let's say that it's a
social media platform, you could still post to it. The trick, though,
is that no one is seeing your posts. Yeah, your
(36:16):
posts are being published, but no one actually gets to
see them. That way, the entity that's making these posts
remains unaware that their work is purposefully being withheld from
the intended audience, and then they're just left to wonder
why the heck their traffic stats are in serious decline.
Tyler and other site owners and operators began to answer
(36:38):
questions from Googlers, So this conversation was really Googler's asking
the site owners questions, and the owners tried to bring
up their concerns about how changes to Google's approach had
disproportionately affected their traffic stats. I recommend reading Tyler's full article.
It is upsetting, it is sad, and as someone who
worked for a company that was really heavily depend on
(37:00):
ranking well and search results, I feel his frustration quite keenly.
Over in Russia, the government there is trying to force
Google to reinstate certain YouTube channels, particularly state backed channels
that are frequently associated with distributing misinformation, particularly about the
war in Ukraine. YouTube has banned those channels for violations
(37:22):
of the company's various policies, and Russia says that's just
not cool. So Russia levy to fine against Google, which
is currently standing at two undecillian rubles. Now I'm pulling
this from John Broadkin's article in Ours Technica. It's titled
Russia finds Google an impossible amount in attempt to end
(37:44):
YouTube bans. So if you're wondering how much is two
Undecillian rubles if you convert it to dollars, well, Broadkin
has kindly done that conversion for us, so it's roughly
the number twenty followed by thirty three zeros in dollars. Like,
it's more money than the world's gross domestic product. And
(38:08):
to say that Google will not pay this is of
course obvious because nobody could pay it. No one, not
the entire world, has that much money. So it's not
really a real fine. It's a message to Google saying, hey,
give us our YouTube channels back. We mean it. Or rather,
it's a fine that really had a stipulation attached to it.
So originally this fine was much less than that. But
(38:30):
if Google refused to reverse its decision to ban the
channels and to pay the fine, that fine would double
after a given amount of time. So have you ever
played the game where you take like two pennies and
then you double it to see how many times doubling
it would take before you're dealing with a good deal
of money. Like if you took two pennies and you
doubled it for eight rounds, you would end up with
(38:52):
two dollars fifty six cents in pennies. If you did
it for sixteen rounds, you've got six hundred and fifty
five dollars and thirty six cents also in pennies. If
you did it for thirty two rounds, you would have
nearly forty three million dollars in pennies. Doubling gets really big,
really fast, so no wonder that fine is now ginormously huge.
(39:16):
Google's response, in a large part over the course of
this dispute has essentially been to pull up Stakes out
of Russia and to shut down most operations there, including
like relocating people who work for Google to leave Russia
and work in other offices. So I suspect we're not
going to see much movement on Google's behalf here. Okay,
I've got a few recommendations for reading for all of y'all,
(39:38):
in addition to the articles that I've already mentioned in
this episode. First up is Eric Berger's piece in Ours Technica.
It is titled what is Happening with Boeing's Starliner Spacecraft?
As we have covered on tech Stuff, the star Liner
has had a really bumpy go of it. A Starliner
spacecraft successfully delivered two astronauts to the International Space Station,
(40:00):
but on the way there, the spacecraft experienced problems with
its thruster systems, and ultimately NASA decided to return the
star Liner to Earth without the astronauts aboard it. They
will have to come back home aboard a SpaceX spacecraft.
At a debriefing designed to address the star Liner as
it had returned to Earth, the two reps expected from
(40:22):
Boeing didn't show up. The company has remained silent about
the star Liner in the weeks following the landing. Two
months have gone by and we haven't really received any
updates on it. Berger explains that, you know, here's what
we know, here's what folks suspect about the spacecraft moving
forward or maybe not moving forward, and so that one's
(40:45):
well worth a read. The other piece that I recommend
reading is Arianna Bindmann's piece in sf gait. It is
titled a Lot of Demoralized People. Ghost jobs are wreaking
havoc on tech workers. So this is about how companies
in general, but particularly in the tech space, will often
post jobs to various career sites, but those jobs aren't
(41:08):
really available now. It may be a case that the
company has already filled the positions in question. It may
be a case that they were never looking to hire
in the first place. And the article explains why companies
do this, and here's a spoiler alert, it's for awful,
cruel and selfish reasons. I highly recommend reading the article
(41:29):
if you have need of being really angry at corporations today.
This is a great way to get very mad at corporations.
It's also just a very good read in general. It's
also very important for anybody who might be in the
job market. If you're someone who has experienced the very
rough setback of having sent out countless resumes only to
hear nothing, this practice might be playing a part in
(41:53):
that this process of posting fake jobs to sites. Understanding
why that happens can also help explain maybe your lack
of traction if you've been job hunting and nothing's come back,
because that can be really demoralizing. It can really add
to stress and depression if you're constantly putting yourself out
(42:13):
there but you're not hearing anything back. Well, part of
the reason of that might be that the companies you're
applying to didn't actually have jobs available in the first place,
and this article explains why, so check that out as well.
That's it for me. I hope all of you had
a fantastic week. For those of you who were out
there trick or treating with the kiddo's or whatever, I
hope you had a great time. Enjoy some mounds and
(42:37):
almond joy on my behalf, and I'll talk to you
again really soon. Tech Stuff is an iHeartRadio production. For
more podcasts from iHeartRadio, visit the iHeartRadio app, Apple Podcasts,
or wherever you listen to your favorite shows. You
Welcome to tech Stuff, a production from iHeartRadio. Hey there,
and welcome to tech Stuff. I'm your host, John van Strickland.
I'm an executive producer with iHeartRadio. And how the tech
are you. It is time for the tech news for
the week ending on November one, twenty twenty four. And
(00:26):
here in the United States, we are rapidly approaching election day,
which is Tuesday, November fifth. Now, for those of y'all
eligible to vote in the United States but who have
not yet done so, I urge you to get out
there and vote. Do your research, look into candidates and
the issues in your area so that you can make
an informed decision, and go deeper than just sites that
(00:48):
are run by politically aligned outlets. Look for information that's
as objective as possible. It's very easy to fall into
a trap of misinformation or bias, and that goes for
any side. I'm not picking specific sides here, it's true
across the board, but for goodness sakes, get out and vote.
It's one of the most important components in maintaining a democracy,
(01:11):
and the very act of voting is under assault these days.
And I don't know about you, but I would rather
continue to have a say, even if it's just a
tiny say in how the country I live in operates.
Now that we have that out of the way, let's
get to a whole bunch of tech stories that relate
to the election, because there are tons of those this week,
which is no big surprise. First up, some voters in
(01:34):
Pennsylvania received a strange text message this past weekend stating
that they had voted in the November fifth election, even
though they hadn't done that yet. Now, obviously that could
stir up some pretty intense confusion and negative emotions. If
you received a message saying that you had already voted
(01:55):
and you had not already voted, clearly you would suspect
hanky and panky going on. Now, in this case, it
wasn't any sort of hanky panky at the actual polling locations.
It wasn't like the state department messed up. Instead, it
was with an organization called All Vote, and All Vote
had previously gotten into hot water for texting registered voters
(02:18):
to alert them that they were not actually registered to vote,
when in fact they were. County officials in different regions
of Pennsylvania have warned that that particular approach was a
hoax designed to gather personal information from people that would
later be exploited against those people. Max Marvin and Kate
Bernard reported on this issue in the Philadelphia Inquirer in
(02:41):
a piece titled Pennsylvania residents got phony texts claiming they
had voted already Ignore them, officials say so. They also
quoted a contracted spokesperson for All Vote named Charlotte Climber.
Climber has claimed that this is all just an innocent
but incredibly dumb and careless mistake made when All Vote
(03:03):
employees entered some sort of typo into their system. So
apparently the original idea was that the organization would text
folks who had voted in the twenty twenty two election
and direct them to voting resources for the twenty twenty
four election. Now, why All Vote felt the need to
inform people that the vote they had cast two years
(03:27):
ago had in fact been counted is a bit beyond me.
That ship has sailed Like I guess the message would be,
you voted in the election, meaning the twenty twenty two election.
Here are some resources for you. But because it left
off the twenty twenty two part, it seemed to indicate
that people had already voted in this election. Now, that
seems to me like that messaging was unnecessary in the
(03:50):
first place. And if that messaging had been left out,
you know, the whole you had voted in the election bit,
there'd be nothing to talk about with this story. It
would just be a text message saying, hey, here's some
resource for you for voting in this election, and then
there'd be nothing to talk about. Climber has said the
founders of All Vote are progressives, but declined to give
more information about them, saying that there's a concern for
(04:12):
their safety and that people who might be hard right
leaning voters might do them harm. This is where I
say that while I consider myself a progressive, so I
would consider myself aligned with the people who are at
least alleged to be behind All Vote. I also think
All Votes operations are not great like they're not done properly.
(04:39):
They were poorly thought out and executed, and if the
owners of all Vote are actually sincere, they're likely to
cause more harm to their cause than to help it
through these kinds of operations. Unless this was an actual
attempt to confuse and discourage people in say hard right counties,
from voting, in which case that's just despicable, Like, I
(05:01):
don't care what side you're for. At least according to
this spokesperson, the people who own all vote believe in
the same things I generally believe in politically, I don't
believe in discouraging people through deception so that they don't vote.
I think that's reprehensible. Whether or not that's what was
going on, I don't know. It could be that this
(05:22):
really was an innocent mistake and just a poorly thought
out campaign in the first place. Either way, a great
way to use technology to discourage the participation of the
democratic process. Speaking of unintentional tech issues affecting the election,
less talk about Colorado. So it seems that the Colorado
Department of State committed a major booboo, or, as Todd
(05:44):
Feathers of Gizmoto called it, a bit of an uh oh.
This comes from an article written by Todd Feathers. It's
titled Colorado agency improperly posted passwords for its election system online. Yeah,
that's not good. So what actually happened? Well, the Department
of State had a link to a spreadsheet on its website,
(06:05):
and the spreadsheet included some hidden tabs, so upon casual glance,
there's nothing hinky going on. However, you can just choose
an option to unhide tabs, in which case one of
those tabs contained passwords for the state's voting machines. Yikes,
I hear you say. Now, it's bad. There's no sugar
(06:27):
coating it. That's bad. That's a terrible breach of security etiquette.
It is, not, however, catastrophic. Why Well, for one thing,
these voting machines actually require two passwords. It's kind of
like one of those things you see in movies where
you have to have two keys to activate the nuclear
arsenal or whatever, and you have to have two people
(06:48):
across the room from each other insert and turned the
keys simultaneously to activate it. Well, in this case, the
voting machines require two separate passwords, and those passwords are
not collected together. So while a set of passwords were
available according to the spreadsheet, the other set was not available.
So even if you had access the spreadsheet and revealed
(07:09):
the tabs and you wrote down all the passwords, you
would still only have one half of the password component.
But that's not at all like in order to actually
use these passwords. You would need to have physical access
to the voting machines themselves. Now, in order to do that,
you would need credentials to be able to get access
to the areas where the voting machines are stored. They're
(07:31):
also under a constant video surveillance, so you'd have to
deal with that as well. Plus, Colorado uses paper ballots,
so there's a physical paper trail of every vote cast.
Voters verify that their ballots are correct before they cast
their vote, so even if someone did access the machine
somehow and messed with them, the paper trail would remain
(07:54):
a resource for verifying the votes that had been cast.
So while the mistake is a giant whoopsie and absolutely
should not have happened, the voting process itself is still reliable.
Of course, that's not stopping some people from using this
to bring into question the results of any votes that
are made in the state. In other words, this is
(08:15):
an opportunity to undermine confidence in the results of the
election ahead of time, so that no matter who wins,
you can have some wiggle room in case your side
didn't win. If your side did win, you say, hey,
look the system worked, and if your side didn't win.
You're like, this is because our system was compromised. Now,
I'm here to remind y'all that for democracy to fail,
(08:36):
you don't actually have to prove that the system is broken.
All you have to do is convince enough people that
the system is broken to reduce their confidence in the system,
and then democracy dies. That's how that works. Reporter Barbara
Ortute has a piece in AP News that's titled report
says crowdsourced fact checks on X failed to a dress
(09:00):
flood of US election misinformation. Now, I think that comes
as a surprise to absolutely no one. Before Elon Musk
carried us SYNC into the HQ of what was then Twitter,
I maintained that was foreshadowing that he would eventually sink
the company, which hasn't happened yet. But goodness gracious, I mean,
I think the band is playing while the Titanic is sinking. Anyway,
(09:23):
the site had content moderators, and they did stuff like
act on tweets that were spreading misinformation, like they could
label tweets as saying this is inaccurate or this is misinformation,
whatever it may be. Those days, however, are pretty much gone.
X relies very heavily on the user community to flag
posts that contain misinformation. You know, these are folks who
(09:45):
are not necessarily trained to do that, or it might
include people who are actually interested in spreading misinformation to
start with. Now, to be clear, this community based tool
existed before Elon Musk took over Twitter. This is not
like Twitter did this in instead of content moderation. Now,
originally this tool was called birdwatch. But while it used
(10:05):
to be a supplemental way for Twitter to monitor and
moderate content, it's now taking a more prominent role because
content moderation is not really a big thing over at
X these days, there aren't that many people left to
actually do it. The Center for Countering Digital Hate took
(10:26):
a look at the community notes tool on X and
found that seventy four percent of the tweets the posts
that they sampled had community notes that were inaccurate or inadequate.
The sample size was two hundred and eighty three posts
that had community notes attached to them. That means that
(10:46):
two hundred and nine of those posts were not accurate.
Now X maintains that the community notes feature is a
useful tool and has pointed to other academic research, saying
that it's a trustworthy source of information. So this is
not to say that definitively the tool is broken or
not performing as intended, but that this one nonprofit group
(11:09):
says that that's the case. For more information and a
more thorough look at all this, I recommend reading an
article by Ashley Bellinger on Ours Technica. It's titled Toxic
X users sabotage community notes that could derail disinfo report says. Okay,
while we're on the subject of Elon Musk, which is
(11:30):
tangential to X, I suppose in PRS Bobby Allen has
a piece titled and Elon Musk backed political group is
posting fake Kamala Harris ads on Facebook. So yeah, we've
got more misinformation and election news headed your way. My
apologies that we knew this was going to happen. I mean,
I've been talking about for months. But in this case,
(11:52):
Elon Musk has provided significant funds to a group called
Building America's Future. This group, in turn, has operated a
Facebook account that is called Progress twenty twenty eight, and
this account has published campaign ads that look like they're
from Kamala Harris's campaign, but in fact they are purposefully misleading,
(12:14):
both to trick the viewer into thinking that the ads
are actually from Kamala Harris's campaign, and also by misrepresenting
Harris's position on various topics. Now, according to the First Amendment,
there's nothing illegal going on here. There's nothing that says
you have to be truthful when you're advertising with political speech. Now, sure,
(12:35):
you got to be truthful when it comes to advertising
products and services. For example, if I were to endorse
something but it turned out I never actually used whatever
it was I was endorsing, I could potentially face a
pretty massive fine for that. Like I could get hit
hard for doing that, My company could get hit hard
for doing that. But you know, when it comes to
(12:56):
informing people about, you know, where politicians stand on specific topics,
truth just isn't a requirement here in the US. Further,
using the name Progress twenty twenty eight for the group
further obvious skates who's actually behind the campaign. So it
would be easy for someone to believe this is a
real ad from the Harris side. And it's interesting because
(13:17):
there are limitations for this kind of thing, I mean,
the use of deep fakes, you know, where you're making
it seem like someone is saying or doing something that
they actually did not do. That could be a violation
of policies. But an ad that ultimately does a similar
thing but without the use of deep fakes, that's fine,
which just makes me curious where the line is. Okay,
(13:38):
we've got tons more information about technology, including within the elections,
but also other stuff. But first, let's take a quick break. Okay,
we're back. Meta held its third quarter earnings call this week,
(14:01):
and CNBC's Jonathan Vanian has a piece all about the
Reality Labs part of that call, which was just one part.
There were there were more things to talk about than
just Reality Labs. But you might remember that Reality Labs
focuses on all things mixed reality, so that includes virtual
reality and augmented reality as well as the metaverse. You know,
(14:23):
this supposed future of our computational approach, and you know,
this has really served as a great way for Meta
to spend tons of money while having to assure investors
that the metaverse is very much going to happen and
it will be the future of computing, and that's something
that stakeholders remain skeptical about. I think for good reason.
(14:45):
That's my personal opinion. I remain unconvinced that the metaverse
is going to be a significant thing. I really doubt
it's going to be the new means of interacting with computing.
I just don't think that's true, but it could be wrong.
So anyway, the call revealed that Reality Labs had a
twenty nine percent increase in revenue, reaching two hundred and
(15:05):
seventy million dollars this past quarter. That's no small shakes.
Two hundred seventy million dollars is a lot of money.
Keep in mind, Reality Labs most of the money they
make is through product like the Meta VR headsets or
the augmented Reality ray bands that Meta has co branded
with ray Ban. Those are the source for revenue in
(15:28):
the most for the most part, for the Reality Labs division,
and two hundred and seventy million dollars that's a lot
of money, but it's not as much as analysts had
been hoping for. They had estimated that Meta would hit
around three hundred and ten million dollars in revenue from
the Reality Labs unit. Further, the division as a whole
chalked up an overall operating loss. So yeah, it made
(15:51):
two hundred and seventy million dollars, but it's spent way
way more than that. How much more? Well, the operating
loss is at four point four billion with a B dollars,
So yeah, you make two hundred and seventy million, but
you spent more than four point four billion. Yikes. Zuckerberg
continues to insist that vr AR and the metaverse are
(16:15):
the future, and he might be correct. Right, just because
it doesn't appeal to me and doesn't really seem to
fit with my concept of what computing really is all about,
that doesn't mean anything at all. I'm just one person,
but I've yet to see a demonstration of this technology
that convinces me that the pros actually outweigh the cons.
(16:36):
I mean, you know, feeling immersed is amazing, don't get
me wrong. Like, if you can experience a really immersive
presentation of some sort, that can be really fun and entertaining.
But wearing a headset for more than, you know, like
half an hour is a pretty hard sell for me,
as well as for a lot of other people. And
you know, people who like me are prone to motion sickness.
(16:59):
If the is not done really, really well, it's a
super hard sell. It doesn't matter how good the experience
is right, like, sometimes our tolerance is pretty low, and
I cannot imagine using this kind of technology to do
work right. Maybe leisure activities, but work is hard to
(17:20):
imagine in this particular context. Now, it could very well
be I'm just suffering under the problems of a very
limited imagination and that this really is the future. I
just remain very skeptical. Jason Kebler of four oh four
Media has a piece that's got a great headline. The
headline is Zuckerberg, the AI slop will continue until morale improves.
(17:44):
That's a great play on a popular joke that's typically
about flogging anyway. As you might suspect, Kebler is not
a huge fan of AI generated content flooding various social platforms,
but he quotes Zuckerberg who said during that earnings call quote,
I think we're going to add a whole new category
(18:05):
of content, which is AI generated or AI summarized content,
or kind of existing content pulled together by AI in
some way. And I think that's going to be just
very exciting for Facebook and Instagram and maybe threads or
other kind of feed experiences over time. End quote. So
that's all direct quote, including like the verbal stumbling now
(18:28):
in full disclosure, I actually don't mind using AI to
summarize something that I have already read in order to
like refresh my memory of the most important points, you know,
kind of like using it like a note taking app.
I don't mind doing that, but I do think it's
still vital to actually read the content first and not
just rely on an AI generated summary, because sometimes those
(18:50):
are not accurate or they misidentify what the most important
points are. I have less than zero interest in having
AI generated content populate my social media feeds as it stands.
I hate that a recommendation algorithm decides which posts I
see and when I see them, because I often miss
(19:10):
out on important, timely stuff because Meta thinks that I
should see that Facebook post later on, maybe sometimes weeks
later after it's been posted. Like I posted a picture
of me having voted. I believe that was on October
seventeenth or so. So anyway, it was not long after Georgia,
my area of Georgia had started doing early voting, and
(19:31):
I was still getting people just seeing that picture for
the first time and liking it this week, more than
a week after I had done it. So that tells
me that the recommendation algorithm is not serving this up
in a timely manner. So I don't want the stuff
that my friends post to be buried even further under
(19:52):
AI generated crap. You know, I'm tired of not being
able to see the things that my friends post. I mean,
there are times where someone appears to disappear from my
life entirely. But it's not that they're gone and they're
not posting or anything. It's that Meta has just decided
that I don't need to see their updates unless I
actually go to their page, and even then I'm not
(20:13):
necessarily going to see it in chronological order. And to
think that even this limited engagement could be limited even
more with AI generated junk is really upsetting. So is
Meta just trying to sever relationships between real life friends. No,
I don't think that's true. I don't think that's what
Meta's goal is. What the company is actually trying to do,
(20:36):
is what it's been doing for years, is trying to
keep you on its various platforms for as long as
possible in order to serve you more ads and to
make more money. And if your friends just aren't enough
to keep you glued to Facebook or you know whatever, Well,
by golly, we'll just have the bot's slave away and
the content minds to create stuff that does keep you there,
(20:59):
which is pretty gross. Oh and while Meta is looking
at a more than four billion dollar operating loss on
the reality Labs division, it's also looking to spend even
more money, largely on the AI side of things. So
according to CNBC, Meta plans to raise the low end
of its capital expenditures guidance for twenty twenty four up
(21:21):
to thirty eight billion dollars, So that's up a full
billion dollars from what it was before. Earlier, the low
end for expenditures for this year was going to be
thirty seven billion. Now the top end is forty billion,
which remains unchanged from earlier in the year. So really
the margin has just decreased, right, It's going to be
somewhere between thirty eight and forty billion dollars, but it's
(21:43):
definitely not thirty seven that's too little. Zuckerberg says next
year will be even more dramatic, which kind of is
in line with what he's been saying for the last
couple of years. And as I have said repeatedly on
tech stuff, AI is expensive and running AI operations in
the cloud costs a whole whole lot of money. You
need a lot of data centers, you need specialized processors.
(22:05):
This racks up the cost of operation very quickly. It's
one of the reasons that I actually favor having AI
processing capabilities built into chips that are in consumer devices
and moving at least some or hopefully all of that
AI processing to the end device, because one that speeds
things up considerably because it's just handling your calculations, not
(22:28):
everybody's two. It vastly improves security and privacy because you're
not constantly beaming your information to some server farms somewhere.
As you engage with an AI application, it's not being
held in some server farm. It's not being used to
train the next generation of large language models. It's just
you interacting with your device. So I much prefer that
(22:52):
approach to AI interactions than the cloud based ones. But
that's really a topic for another time now. Like reality labs,
the AI costs are something Zuckerberg doesn't really shy away from.
He just wades right into it. He says it's a
necessary cost in order to establish the infrastructure for future platforms,
(23:12):
and again he might be right, like, maybe this is
the future and he's just getting the work done early,
and it's very painful because it's very expensive and it's
not paying off right away, but that in the long
run it sets the groundwork that allows Meta to prosper.
But in the meantime, the market has been a little critical.
Shortly after the earnings call, in after market trading, the
(23:35):
stock price for Meta took a bit of a dip.
So it seems to me that investors are getting a
little anxious and tired of these reports about the company
spending so much money on reality labs and AI when
the payoff for those investments is pretty far down the road. Personally,
(23:57):
I think in general that preparing for the long term
is a better approach than just focusing on the short term. So,
you know, I feel conflicted about this because I do
think that if you're going to really invest in a
company's future, being willing to do that and know that
you're not going to see returns right away takes a
(24:17):
lot of bold decision making and ultimately can be the
best thing for the company. On the flip side, I
remain unconvinced that the metaverse and AI is really going
to give Meta the ability to stand apart from other
competitors in the space, or that the company has established
that there's a future in that particular type of computing
in the first place. So double edged sword kind of thing.
(24:40):
But I would actually like to see more companies take
a firm stance on yes, this is a painful decision
in the sense that you know, we're investing money in
something that's not going to be an immediate payoff, but
down the road it's going to be the best thing
for our company. I am really tired of seeing companies
that almost sacrifice everything for the purpose of getting short
(25:02):
term gains, because ultimately, I think that ends up corrupting
companies from the inside, and they ultimately crumble under their
own weight once they reach a point where you just
can't do that anymore. Okay, enough about that, But one
other thing that Zuckerbird brought up during that earnings call
was that Threads, the Meta operated alternative to Twitter, is
(25:23):
up to nearly two hundred and seventy five million monthly visitors.
That is a one hundred and seventy five percent increase
from last year, and he also said the app is
seeing more than one million new users sign up every
single day. Now, that, along with a general migration away
from X formerly known as Twitter, means the gap between
(25:46):
the two services is beginning to narrow. Analysts from Sensor
Tower estimate that X has somewhere in the neighborhood of
three hundred and eighteen million users. Now, I say estimate
because X is a private company, and as a private company,
they don't have to share that kind of information. Threads
is somewhere around two hundred and seventy five million monthly users.
(26:07):
So the growth of Threads, along with folks who are
ditching X, has changed the landscape quite a bit. Meta's
CFO Chief financial Officer Susan Lee has said that the
company doesn't expect Threads to be a significant revenue driver
in twenty twenty five. Rather, the focus will be on
growing the platform further before really exploiting the heck out
(26:28):
of it. Now, if I were a betting man, I
would say that Meta will eventually sprinkle some of its
targeted advertising ferry dust on Threads, and it will probably
happen in the not too distant future, and it will
increase Meta's already significant hold in the online advertising space. Meanwhile,
I expect X's decline to continue simply because Elon Musk
(26:51):
cannot help himself and consistently makes decisions that drive away
many users of X to other platforms include Threads and
blue Sky and mast it on. I think the one
advantage X has is that no other singular microblogging platform
has the dominance that Twitter once had, So either users
(27:14):
are adopting multiple platforms and kind of spreading their time
across them, or they're having to choose one of several
different offerings and just be satisfied with the fact that
they're not going to see everything they used to see
because other people have gone to competing services. So that's
kind of my view of where things are right now. Okay,
we're going to take another quick break, but I still
(27:35):
have plenty of tech news. This week was shock full, y'all,
so let's take a quick break. We'll be back with
some more headlines. Kylie Robson of The Verge has a
piece titled open AI's search engine is now live in
(27:57):
Chat GPT, which I'm sure is for concern for some
folks over at Google. Now. To be clear, this feature
is currently only available to paid subscribers of Chat GPT.
It will later roll out to other users, like, you know,
enterprise users or people who are working in educational institutions.
(28:17):
As Robison explains, the feature is integrated directly into the
chat gpt experience, so it's not like a separate tool
or tab or something. Users when they're putting in a
query and conversing with the chat gpt bot will sometimes
receive suggestions for various websites that relate to whatever the
(28:41):
topic of discussion is, so they'll get suggestions for resources
that they can use to further answer their questions. Robison
points out that chat GPT's search results do not contain
promoted results, so that might be a big advantage over
folks who previously had used Google. You can, and also,
by the way, directly ask chat gpt to give you
(29:04):
web results, so you can ask it to give you
those those websites from the beginning and just treat it
like it's a search engine. And by not including promoted results,
that's a shot across the bow at Google, because that's
Google's bread and butter. I'm guessing anyone who has used
Google for like the last decade has noticed that sponsored
(29:25):
results appear at the top of search results. Lists now.
In the very old days, the general expectation was that
the most relevant, highest quality search results would appear at
the very top of Google's results page. But for the
last several years, the very top of that list is
frequently dominated with paid ads that lock in those spots,
(29:48):
And even though Google does tag those search results as
being promoted search results, some folks still bristle at how
this feels like preferential treatment for what is often thought
of as a scholarly like ibbe. As if you went
to the library and you wanted to look up books
that were about a specific topic, and you found out
that the results of all the books in the library
(30:10):
that do relate to that topic have been ordered not
by author name or whatever, but by way of who
has paid the library to have their book promoted the most.
That's not cool. Open ai claims it does not intend
on selling advertising the way Google does, but that does
raise questions as to how AI will generate revenue from
(30:33):
free users who use this tool, because presumably free users
will be able to interact with this in the not
too distant future. Another question that remains is how open
ai will navigate the choppy waters when it comes to
referencing or pulling from various resources online. Most media companies
would very much prefer if you were to go to
(30:53):
their website and read stuff on their own web pages.
That's one of the many reasons I actually recommend folks
check out the articles that I reference in these episodes. BENJ.
Edwards of Ours Technica, as an article titled Google CEO says,
over twenty five percent of new Google code is generated
by AI. That's really incredible that more than one quarter
(31:15):
of the code generated within Google is coming from AI tools. Now,
according to Google executives, all these operations are still ultimately
overseen by actual human beings. AI might generate some of
that code, but humans review and then approve that code
before it can be deployed. Now, the purpose of this
structure is to streamline coding and speed things up, presumably
(31:38):
while also cutting down on the need for quite so
many people coding for Google in the first place. And
as Edwards explains, not everyone is super happy about how
AI is taking such a prominent role in coding. There
are critics who worry that relying so heavily on AI
means we could end up with apps and code that
has more errors in it more vulnerabilities interduced into the
(32:00):
code that are difficult to detect, meaning that anyone deploying
that code could be creating opportunities for hackers, and maybe
the hackers are using their own AI tools to identify
and then exploit those vulnerabilities. In fact, I think that's
just a guarantee. I think AI coding and AI hacking
that's the reality we face now. It's going to be
(32:21):
a seesaw like relationship moving forward. I recommend reading BENJ.
Edwards article in full for a deeper look into the story,
including how some studies have shown that folks who have
used AI to help code will simultaneously end up believing
that their code is superior to other people's code because
they used AI tools to make it, while also introducing
(32:43):
more bugs into their own work. So it's kind of
like creating this false sense of superiority or our expectation.
Jonathan M. Gitlin, also of ours Tetnica, has another Google piece.
This one is titled Generative AI is Coming to Google Maps,
Google Earth Ways and Yeah. It's all about how Google
(33:04):
is incorporating more AI tools into its various map apps
to give users more features. Some of this was news
to me, In fact, quite a bit of news to me,
even though apparently some of these features have been around
for a while. So for example, Getland quotes Google's VP
and general manager of GEO. That's Chris Phillips, who said, quote,
(33:26):
think of features like lens and maps. When you're on
a street corner, you can lift up your phone and look,
and through your camera view you can actually see. We
laid places on top of your view so you can
see a business, Is it open, what are the ratings
for it? Is it busy? You can even see businesses
that are out of your line of sight. End quote.
(33:47):
I didn't know about that, although I have talked about
that very use case and augmented reality in the past.
But I think it's pretty cool, particularly if you're out
and about somewhere that's new to you and you want
to find a place that's like open and nottally slammed.
I actually could have used that. Last weekend, me and
my party were in Pennsylvania for the Pennsylvania Renaissance Fair,
(34:07):
which now is over, but if you ever get a chance,
it's a phenomenal renaissance fair. I say that as someone
who used to work at a Renaissance Fair. It's amazing.
But after the fair was done on Sunday, we wanted
to go and eat at a restaurant, and we had
identified which one we wanted to go to, but by
the time we got there, we found out the kitchen
had already closed, and so we were forced to change
(34:28):
our plans and find somewhere else to grab dinner. Having
this tool would have told us like, oh, by the
time you get there, it's going to be too late,
so don't even bother, and that would have saved us
some trouble. Anyway, Google is bringing AI enabled search results
to Maps, so if you just have a general hanker
in for something but you don't know any more details,
you can ask Maps and have a little conversation with
(34:50):
an AI bought to whittle down your options. Check out
Getland's article to learn more about the upcoming features headed
to Google Maps and these other tools. Joshua Tyler of
Giant Freaking Robot has an article titled I attended Google's
Creator Conversation event and it turned into a funeral. Yikes.
All right, so this is about an event that Google
(35:11):
held this week, and Tyler's description of it is pretty
darn grim. He talks about receiving a tour of the
Google campus, and it was a tour of a campus
that was almost completely empty, although the event was happening
in the middle of a work day. He describes getting
not one but two security badges before being allowed into
a secure building, only to have no one actually look
(35:33):
at or check his credentials once that time came. According
to Tyler, the attendees of this event consisted of folks
who worked on independent websites that have been shadow band
due to Google's changes in its approach to generating search results. Now,
in case you're not familiar with the term shadow band,
it means that you're blocking someone or something from your
(35:56):
online service like a forum or in this case, Google
search results, and they don't have any knowledge that they've
been blocked. So the entity is still allowed to make
posts on the platform. Like let's say that it's a
social media platform, you could still post to it. The trick, though,
is that no one is seeing your posts. Yeah, your
(36:16):
posts are being published, but no one actually gets to
see them. That way, the entity that's making these posts
remains unaware that their work is purposefully being withheld from
the intended audience, and then they're just left to wonder
why the heck their traffic stats are in serious decline.
Tyler and other site owners and operators began to answer
(36:38):
questions from Googlers, So this conversation was really Googler's asking
the site owners questions, and the owners tried to bring
up their concerns about how changes to Google's approach had
disproportionately affected their traffic stats. I recommend reading Tyler's full article.
It is upsetting, it is sad, and as someone who
worked for a company that was really heavily depend on
(37:00):
ranking well and search results, I feel his frustration quite keenly.
Over in Russia, the government there is trying to force
Google to reinstate certain YouTube channels, particularly state backed channels
that are frequently associated with distributing misinformation, particularly about the
war in Ukraine. YouTube has banned those channels for violations
(37:22):
of the company's various policies, and Russia says that's just
not cool. So Russia levy to fine against Google, which
is currently standing at two undecillian rubles. Now I'm pulling
this from John Broadkin's article in Ours Technica. It's titled
Russia finds Google an impossible amount in attempt to end
(37:44):
YouTube bans. So if you're wondering how much is two
Undecillian rubles if you convert it to dollars, well, Broadkin
has kindly done that conversion for us, so it's roughly
the number twenty followed by thirty three zeros in dollars. Like,
it's more money than the world's gross domestic product. And
(38:08):
to say that Google will not pay this is of
course obvious because nobody could pay it. No one, not
the entire world, has that much money. So it's not
really a real fine. It's a message to Google saying, hey,
give us our YouTube channels back. We mean it. Or rather,
it's a fine that really had a stipulation attached to it.
So originally this fine was much less than that. But
(38:30):
if Google refused to reverse its decision to ban the
channels and to pay the fine, that fine would double
after a given amount of time. So have you ever
played the game where you take like two pennies and
then you double it to see how many times doubling
it would take before you're dealing with a good deal
of money. Like if you took two pennies and you
doubled it for eight rounds, you would end up with
(38:52):
two dollars fifty six cents in pennies. If you did
it for sixteen rounds, you've got six hundred and fifty
five dollars and thirty six cents also in pennies. If
you did it for thirty two rounds, you would have
nearly forty three million dollars in pennies. Doubling gets really big,
really fast, so no wonder that fine is now ginormously huge.
(39:16):
Google's response, in a large part over the course of
this dispute has essentially been to pull up Stakes out
of Russia and to shut down most operations there, including
like relocating people who work for Google to leave Russia
and work in other offices. So I suspect we're not
going to see much movement on Google's behalf here. Okay,
I've got a few recommendations for reading for all of y'all,
(39:38):
in addition to the articles that I've already mentioned in
this episode. First up is Eric Berger's piece in Ours Technica.
It is titled what is Happening with Boeing's Starliner Spacecraft?
As we have covered on tech Stuff, the star Liner
has had a really bumpy go of it. A Starliner
spacecraft successfully delivered two astronauts to the International Space Station,
(40:00):
but on the way there, the spacecraft experienced problems with
its thruster systems, and ultimately NASA decided to return the
star Liner to Earth without the astronauts aboard it. They
will have to come back home aboard a SpaceX spacecraft.
At a debriefing designed to address the star Liner as
it had returned to Earth, the two reps expected from
(40:22):
Boeing didn't show up. The company has remained silent about
the star Liner in the weeks following the landing. Two
months have gone by and we haven't really received any
updates on it. Berger explains that, you know, here's what
we know, here's what folks suspect about the spacecraft moving
forward or maybe not moving forward, and so that one's
(40:45):
well worth a read. The other piece that I recommend
reading is Arianna Bindmann's piece in sf gait. It is
titled a Lot of Demoralized People. Ghost jobs are wreaking
havoc on tech workers. So this is about how companies
in general, but particularly in the tech space, will often
post jobs to various career sites, but those jobs aren't
(41:08):
really available now. It may be a case that the
company has already filled the positions in question. It may
be a case that they were never looking to hire
in the first place. And the article explains why companies
do this, and here's a spoiler alert, it's for awful,
cruel and selfish reasons. I highly recommend reading the article
(41:29):
if you have need of being really angry at corporations today.
This is a great way to get very mad at corporations.
It's also just a very good read in general. It's
also very important for anybody who might be in the
job market. If you're someone who has experienced the very
rough setback of having sent out countless resumes only to
hear nothing, this practice might be playing a part in
(41:53):
that this process of posting fake jobs to sites. Understanding
why that happens can also help explain maybe your lack
of traction if you've been job hunting and nothing's come back,
because that can be really demoralizing. It can really add
to stress and depression if you're constantly putting yourself out
(42:13):
there but you're not hearing anything back. Well, part of
the reason of that might be that the companies you're
applying to didn't actually have jobs available in the first place,
and this article explains why, so check that out as well.
That's it for me. I hope all of you had
a fantastic week. For those of you who were out
there trick or treating with the kiddo's or whatever, I
hope you had a great time. Enjoy some mounds and
(42:37):
almond joy on my behalf, and I'll talk to you
again really soon. Tech Stuff is an iHeartRadio production. For
more podcasts from iHeartRadio, visit the iHeartRadio app, Apple Podcasts,
or wherever you listen to your favorite shows. You
TechStuff News
Host
Jonathan Strickland
Popular Podcasts
Las Culturistas with Matt Rogers and Bowen Yang
Ding dong! Join your culture consultants, Matt Rogers and Bowen Yang, on an unforgettable journey into the beating heart of CULTURE. Alongside sizzling special guests, they GET INTO the hottest pop-culture moments of the day and the formative cultural experiences that turned them into Culturistas. Produced by the Big Money Players Network and iHeartRadio.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations.
Crime Junkie
If you can never get enough true crime... Congratulations, you’ve found your people.