← The NewsWorthy
Loading summary
Transcript67 lines
- [00:00]
A
Today is Saturday, December 13th. As 2025 comes to a close, we're looking back at one of the biggest drivers of change this year, tech. And counting down the top 10 tech stories, at least according to our guest. He's Brian McCollough, host of the daily podcast Tech Brew Ride Home, covering all of Silicon Valley's news in a 15 minute daily roundup. He's sharing what broke through, what disappointed, what shocked the industry and what all of it could mean for 2026. Let's get into the top tech stories of 2025, from number 10 to the biggest tech story of the year. Welcome. Welcome to the newsworthy special edition Saturday when we sit down with a different expert or celebrity every Saturday to talk about something in the news. Don't forget to tune in every Monday through Friday for our regular episodes where we provide all the day's news in 10 minutes. I'm Erica Mandy. It's now time for today's special edition Saturday. Ryan Mecola, thank you so much for joining us here on the Newsworthy again.
- [00:58]
B
Yeah, thanks for having me back. This is a great annual tradition we're doing here.
- [01:01]
A
I know, I'm loving it. Well, let's just start before we get into the top 10. Let's start big picture. In just a couple of sentences, how would you describe tech in the year 2025?
- [01:12]
B
Back tech is back. It rebounded. All sorts of things hit all time. Highs from Nvidia stock to Bitcoin to things like IPOs starting to happen again. So tech had a very good year. As the year is ending, there are lots of storm clouds on the horizon. I think we'll get to that.
- [01:30]
A
Yes, we will. Before we get into it, what were you thinking about as you were creating your list? Like, what was the criteria for something to make the list?
- [01:39]
B
Things that I think will look back historically as this was the year this broke through or that happened. It's hard not to have made it all about AI. But you'll see AI kind of weaves its way through half the things we're going to talk about.
- [01:52]
A
Yeah, let's go through the top 10 biggest tech stories of the year. And if we can, I'd love to start with number 10 and work our way up to number one. So let's go. Give us number 10.
- [02:02]
B
Number 10 would be smart glasses as a new sort of consumer hardware product category. But again, here's AI weaving its way in already. Because what's happened is this is a product category that is enabled by AI. If suddenly you can, you know, add A camera to glasses, but then also have a connection to your phone and so you can talk to it and then the AI can like talk back to you in your glasses, but then eventually overlay things in your vision. In the real world, multimodal AI is a thing. It has become consumer product, moving from like a niche to things like Meta's Ray Bans selling, we think, tens of millions of units at this point.
- [02:41]
A
Yeah, because we've heard of smart glasses a lot in the past, but you think this year it's actually getting sold and there's new features. That's what's really making it stand out this year.
- [02:50]
B
It's being cheap by just adding the AI stuff, like the Apple Vision Quest. This is different from, you know, full VR stuff. This is just, we'll do the best we can now and make it affordable. But adding the AI added enough of a gloss to it that has made it an actual viable consumer product, I think.
- [03:08]
A
Have you seen them out in the wild?
- [03:10]
B
Out in the wild? No. I mean, I've tried a bunch of them on and I do not use one on a daily basis. So that's an interesting thing. A lot have been sold, we're told. But you're right, I don't see them on the subway.
- [03:20]
A
All right, let's go to story number nine.
- [03:23]
B
This is one of those where I think we'll look back on this year as the year quantum computing got practical. It's sort of like AI, you know, years and years of hype. And then we had the ChatGPT moment. I think for quantum computing, this is the year where it became viable. Real world milestones were hit through hybrid computing, mixing classical supercomputers with quantum processors and researchers successfully applying the tech to actual problems in, like, material science and drug discovery. So we're proving this year in 2025 that it's not theoretical anymore, even if we haven't hit the ChatGPT moment for quantum computing yet.
- [04:00]
A
And how do you attempt to explain what quantum computing is to the average person?
- [04:06]
B
Not going to try that. I'm just going to say that it can do things that our classical computer models would take them tens of thousands of years to compute. When we crack this quantum computing, it can happen in seconds. So it could do things like break cryptography or, I don't know, not only cure cancer, but maybe allow us to do cold fusion and stuff like that. Imagine like layering AI on top of computing that is orders and orders of magnitude faster. And you can see why people are excited about it.
- [04:41]
A
All right. Number eight.
- [04:42]
B
Number eight is another things coalescing and this is sort of like the breakthrough year. Self driving cars, I think we'll look back on this is the year that self driving cars hit scale. Things like Waymo, but also Uber, et cetera. We knew that once they cracked it in one city then it's just like a franchise model in like restaurants or something. You just open up in another city and another city. And we're seeing that there are close to a dozen cities now that have self driving cars in them. There's close to half a dozen that have. You can actually pick up your phone and instead of calling for an Uber driver, you get a self driving car to come pick you up. And in fact we've seen data that in cities like Austin, Texas, in San Francisco, where these self driving ride hailing apps are available already as much as 15 to 20% of the ride hailing are now being driven by cars that have no human driver in them. So extrapolate that out. If a year from now it's 30 cities, then 150 and as much as 20 to 30% of the ride hailing is going to be done by self driving. You can see that this is going to be something that everyone's going to experience quite soon.
- [05:53]
A
Despite some controversial stories that have come out, a couple have hit some pets or animals, but that's not necessarily slowing it down.
- [06:01]
B
It's not slowing it down because once people get comfortable with it, it's theoretically cheaper, it's easier. And listen, there is data that came out in the last month that proved over millions of miles driven, that it was Waymo data that proved that Waymo cars are safer than humans. There was a New York Times op ed that a neurosurgeon said we need to implement this soon as possible as a public health revolution. If we could eliminate the 50,000 people that die in car accidents a year, or eliminate or even scale it down by a third or whatever, like it's something that would be a public health miracle.
- [06:40]
A
Do you think my 4 year old will have to learn how to drive ever?
- [06:43]
B
I actually don't think so.
- [06:44]
A
That's what my husband and I have talked about before. Like he may never actually learn how to drive. We'll see what happens.
- [06:49]
B
It's going to be something that a decade from now you'll be like, I can't believe we used to do that. And that's really the scale that I think, the timeline that I think we're hitting. A decade from now it'll be common.
- [07:01]
A
And I mean, once it scales, there are going to be stories of problems because there's just. You look at the numbers for every.
- [07:07]
B
Bodega cat that gets hit. You haven't heard the story of the, what, a thousand people that were in car accidents today?
- [07:14]
A
Right. Okay, number seven, Slop.
- [07:18]
B
AI slop is the word of the year because of the AI revolution, because of things like, you know, Sora and all these tools that make it so easy to create not only AI pictures, but now AI videos. First the social media corner of the Internet, but now, gradually all corners of the Internet are being taken over by a deluge of again, what some people would call low quality AI generated content to, I don't know, just content in general. And it's been interesting as it's taking over. On the one hand, a lot of users and people are actively revolting against this sort of synthetic slop. But then at the same time, why are places like meta and OpenAI creating essentially social networks for this AI content? It's because they can see in their own data, and you can see by the view counts that people like watching this stuff. If you do it right, yeah.
- [08:13]
A
What do you think the impact of quote unquote, AI slop is now and or next?
- [08:21]
B
I am not in favor of this. I take no position on this. But I think it's inevitable that it will essentially maybe not take over social media, but subsume it. I just think because it's so easy to do, I think it's an inevitability, and we're only in like inning one or two of how easy it is to do so. If you're a creator with 10 million followers or something, if you can create the best video in the world of you standing in front of the Grand Canyon and you can do it in 30 seconds, it's still your content and your ideas, but why won't you do that?
- [08:58]
A
But when it comes coming from news, how concerned do you think people need to be about understanding what is real and what is not?
- [09:05]
B
Obviously we should all be extremely concerned about that. I just wonder the degree to which. Again, because there is essentially this very strong backlash against it and very strong skepticism. I don't know that it'll be like you flip a switch and everybody's fooled by everything. I think it'll be more of a sort of prolonged back and forth between people fighting it, people hating it, and then also because again, it's just cheaper and easier. That's what I think is inevitable about it. People are just going to use it.
- [09:34]
A
All right.
- [09:35]
B
Number six, sorry I again, but this is something that didn't happen this year. Agentic AI kind of didn't happen. We've had chatbots and such, but what we were told at the beginning of the year was this is the year where, well, you'll have a agent on your computer that will book a flight for you. Research that slide deck that you have to make for work, research it. But then the next step is it'll create it for you and then it'll edit for you and you'll work with these agents all day. And that kind of hasn't happened yet. We've seen a bunch of agentic stuff come out, but I haven't seen the sort of takeoff and sort of adoption of it that everyone was promising.
- [10:15]
A
Right. It was reported that it's available, but are people actually using it and is it actually capable of doing what it says?
- [10:22]
B
I'm not sure yet if the problem is it's not capable. I just know that it's not being adopted.
- [10:28]
A
Okay, number five, you know what?
- [10:30]
B
I'm looking at this. Oh, except for one thing. It's AI all the way down.
- [10:34]
A
Well, I mean, that is the reality of the year. Right. As much as maybe somebody's sick of hearing about it, it's because it has been the biggest story of the year.
- [10:41]
B
It is with, with different angles to it. So this angle would be sort of the AI arms race. So this would be Deep Seek and Chinese, I don't know if you remember at the beginning.
- [10:52]
A
Yeah, we reported on this.
- [10:53]
B
Yes. So the assumption of American hegemony in AI was shaken by things like Deep Seq, which was created by a Chinese company and that outperformed some of OpenAI's top models and was apparently trained for like 1/100 of the money required to train OpenAI's top models. So on one hand this is, oh, is the US losing its lead? But the second angle to it is that these are open source models where essentially companies put these out and then anyone can adopt them and use them. And I run some of them on my laptop. So this is sort of not just, oh, are the Chinese going to win? Is American tech going to win? It's also, is a closed model versus an open model going to win. And again, if it's cheaper to use and it's good enough, it's going to win out. And that's been proven time and time again in tech history.
- [11:47]
A
And there's been some arguments from like AI ethicists that the AI race, the arms race is what's going to be the failure of society. Like we have the wrong incentives if we're just trying to be bigger and better versus doing it responsibly. What's your take on that?
- [12:05]
B
A lot of people are familiar with Moore's Law, this idea that computing gets faster every couple of years and gets cheaper every couple of years. And like that's the thing that has driven, you know, the tech eating the world of the last 30 years. That is starting to play out to a certain degree in AI. Whether or not this is morally the right thing to do for, you know, the survival of the human race, I mean that's a deeper, larger philosophical debate. I mean there's a reason why in the Dune novels they don't use AI because they fought a war against the AI and we all remember Terminator and all that stuff. So I like.
- [12:44]
A
So we'll see how that plays out in real life. Okay, number four.
- [12:47]
B
Number four would be Nvidia, the first $5 trillion company. Nvidia was the big story of the year by the fact that they make the chips for the AI, People can't get enough of it. They can charge basically anything they want and people gobble up as much as they can. But that is because this idea that all of the major tech platforms are just spending like it's going out of style to provide you with AI, they think you're going to want it, they have to build out these huge data centers to provide it to you is AI. And this has been a debate all year long. I did a bubble. Well, is it a bubble? But is it also propping up the entire US economy? Early in the year I did a big story where I interviewed an economist that was like without this AI spend out from Google, from Amazon, from whomever, they're spending hundreds of billions of dollars a quarter to create these AI data centers. Without that, we probably would already be in a recession. It's essentially like this huge hundreds of billions of dollars a quarter stimulus not from the government, but from private companies that you could argue is propping up the entire economy right now. We don't know if it's a bubble right now, but let's put it this way, what needs to be proven, Is there a return on investment for all of this investment? If there isn't, then this was a one time stimulus. But hey, we saw something like that in the dot com bubble where hundreds of billions of dollars were thrown out of new technology. No one was wrong. It did turn out that this was the technology that took over. It was just that some of it was too soon. And in any bubble, even if on the long arc of history, it is the right investment, the timing of the investment is what matters for whether or not a bubble pops.
- [14:34]
A
So lots to watch moving forward, but this is the year that I think we've seen the most investment for sure.
- [14:39]
B
Even adjusted for inflation, it's the most investment in technology since the late 90s.
- [14:47]
A
Still ahead, we're counting down the final three biggest tech stories of 2025, plus the one emerging technology our guest is most excited to be watching closely in 2026. That and more after a quick break for our sponsors. This episode of the Newsworthy is brought to you by Skims. As the holidays near, I try to do even little things around the house to create a little bit of that magical, joyful vibe. And having super cozy, soft and festive pajamas is definitely one of those little things that can make a big difference. I appreciate that. I don't feel like I have to hide away when we take a spontaneous family picture early in the morning. And really, holidays or not, the Skim Sleep set is my new go to just because of the fabric. It is so soft and nice on my skin and it's cozy without being too warm. I think it actually helps me sleep better. And with women's, men's, kids, and even pet pajamas, Skims makes it really easy to check off a lot of people on your holiday list at once. Shop my favorite pajamas@skims.com after you place your order, be sure to let them know we sent you select podcast in the survey and be sure to select our show the Newsworthy in the dropdown menu that follows, click and if you're looking for the perfect gifts for everyone on your list, the Skims Holiday Shop is now open@skims.com the newsworthy is also brought to you by Oneskin. This time of year gets busy, fast travel events, eating more and different foods. It all adds up. So I find myself craving simple rituals that make me feel like I'm actually taking care of myself. And for me, one of those rituals is my one skin routine. Their face moisturizer has become a staple and the new peptide lip mask is a must have for me as well. Now that's why I'm thrilled to tell you that both are part of Oneskin's nightly Rewind gift set. In fact, you can check out all their limited edition holiday sets that help renew skin at the cellular level for a stronger, smoother and more youthful looking complexion. These make great gifts. Of course the Core of all their products is their OS1 peptide proven to target the root cause of things like wrinkles and loss of elasticity. Plus one skin has been certified safe for sensitive skin. Their products are free from over 1500 harsher irritating ingredients. They're dermatologist tested and have been awarded the National Eczema association seal of acceptance. For a limited time, try OneSkin for 15% off using the Code Newsworthy at Oneskin Co Newsworthy. Again, you can try OneSkin for 15% off using the code Newsworthy at Oneskin Co Newsworthy. After your purchase they'll ask you where you heard about them. So please support our show and tell them we sent you. Now back to my conversation with Brian McCollo. I think we're down to the top three. Give us number three.
- [17:20]
B
All right, one that's not AI, but this is kind of tied into tech's good, strong, healthy year. Crypto took over, hit all time highs. You know, you had the Trump administration come in and be crypto friendly and essentially betting on even the smallest coins or things like the prediction markets which are built on top of the blockchain or whatever. Everything took off, everything came back. Bitcoin hit all time highs. Ethereum hit all time highs. All of this stuff you had companies that they became, I think the term for it is crypto treasury companies, where companies would just say, hey, we're going to invest all of our money in buying up this token or that thing. Now hit those highs and in the last two or three months a lot of that is starting to evaporate. Crypto is sort of having a hangover moment right now. Number two, number two is going to roll right into number one, which is what I would call the disappointment of GPT5. So OpenAI being the leader for the last several years in AI, the best models, the frontier models and you know, the move from GPT3 to GPT4 was sort of a phase change, a step change. Like GPT4 was so much better than GPT3 that there was all this hype, well, what is GPT5 going to do? Will they be the only company we need? Because it'll be an all knowing or oracle that can solve every problem and make all the money for you, et cetera, et cetera. It was not that and I'm not saying GPT5 was bad, but it was a sort of ho hum rollout that people were like, yeah, this is good and more people are subscribing to ChatGPT and things like that. But it was not the step Change that people were expecting.
- [19:10]
A
Well, and everything with AI is moving so fast. It's almost like the expectations now are different than they ever used to be. Like we're seeing it move fast. So if the next one isn't moving just as fast, it's surprising.
- [19:24]
B
Hey, guess what? That's a perfect segue into the number one story.
- [19:28]
A
All right, drumroll please. Number one.
- [19:31]
B
Is OpenAI in trouble? Is Nvidia in trouble? Is Google suddenly now the leader in AI? And what does all of that mean? So, just mentioned the sort of botched rollout of GPT5. Google released Gemini 3 and the consensus broadly is that it is superior to GPT5. So what does it mean for OpenAI if they are not the acknowledged leader in AI? Well, they have to raise money. They're a private company. Google is a public company with all the money in the world. Is this a situation to bring it historically back to the dot com era? Netscape was the first dot com company. Netscape did not even survive the dot com era. Is OpenAI going to be the leader that got arrows in its back and then gets killed by better models? The greater resources of other companies? Why is OpenAI behind? They just announced a code red where they're scaling back some of their other ambitious plays and are like, we gotta get our models the best again. Okay, Now Google being ahead is potentially trouble for Nvidia. Why is that? Well, Nvidia makes GPUs. The chips that run AI are graphic processing units as opposed to CPUs. And Google has GPUs, their own specialized chips that like a decade ago, they created for their own internal AI. Well, if Google's model is better, then why don't the anthropics of the world, that startup that you'll hear about tomorrow that has made AI for, I don't know, lawyers or whatever, why won't they use the Google LLMs and then the Google chips, the TPUs? So in the last month, people are suddenly questioning things that we talked about earlier on this list. Is OpenAI still the leader? Is Nvidia still the leader? Or suddenly is the whole AI horse race like completely up in the air?
- [21:30]
A
Right. And I guess if you also think about what are people going to be expected to use at work versus what they might choose to use at home, and what did they become familiar with?
- [21:38]
B
Yeah, and you know that enterprise trusts Google for lots of enterprise stuff. The other thing about Anthropic is, is that they've always been known as the. Getting back to your question of should we do this safely. We're the safe AI company. You know, Sam Altman said a couple months ago that OpenAI was even going to allow you to chat dirty, do erotic chats with ChatGPT. That is not what anthropic does. Anthropic is like, we're going to make this safe, we're going to make this reliable, and we're going to make sure that it is saf for your enterprise to use. And maybe that was the strategic way to do it, where if enterprise is the way to go and enterprise trusts you, then maybe the anthropic and the Google model is the model that's going to win.
- [22:18]
A
I didn't hear from you anything about AI companions, which I know for at least kids that has been a big thing. What's your take on that?
- [22:26]
B
They've had to scale it back incredibly. There's been all sorts of lawsuits about bad things happening to people that trust these chatbots to use them for therapy or all sorts of other reasons. That is something that I think is going to continue to be iffy. I don't expect that to take off in any major way unless it is a thing where people are sort of using it on the down low.
- [22:49]
A
Has there been an impact from tariffs or the DEI pushback that's worth mentioning?
- [22:54]
B
You know what? If I had to do an 11, it would be that essentially that tech has had to deal with the impact of tariffs all year long. The reason I left it off the list is because tech's not alone in that. But yes, navigating this sort of, well, this month, all of a sudden it's going to cost you 30% more to import your iPhone or your GPU or whatever. And then next month that's gone away and then a month later it comes back again and now it's 15%. It's been annoying and it has been a material thing to that tech companies have to deal with. But in the end, it's just been something that people have navigated, but it hasn't really blown anything up.
- [23:33]
A
Which story are you personally excited about? You know, not just as a tech expert and reporter, but just on your own time.
- [23:40]
B
This is another thing that we've expected for years, going back to the Jetsons in the 1950s. But I'm going to go to CES in a couple weeks and my prediction is that this is the year that robotics is the big story coming out of ces. And this is another AI angle because for the first time in my professional life, we are seeing AI home Robots, robots that you could buy for a couple thousand dollars, bring them into your home and you could have them do the dishes for you. I'm not saying that they're going to be perfect. I'm not saying they're going to be adopted overnight. But this is the first time viable consumer home robotics, I think, is becoming a real thing again. It's because AI is allowing them to navigate your home, navigate the real world autonomously. If this was the year that smart glasses became a real consumer product, I think this is the year that we're going to start to see actual consumer home robotics in a meaningful way.
- [24:38]
A
And that leads me into the next question, which is looking ahead, what else are you watching for in 2026 if.
- [24:45]
B
The AI bubble bursts? Because I'm going to posit that it is a bubble, it's just a matter of when it bursts and to what degree it does burst. If it happens next year, that'll be the biggest story of the year. Because so much of everything is built around all of these investments, the capex spend, the hype, et cetera. We didn't even get into these sort of circular investments between OpenAI, Nvidia and all these other. If, if OpenAI crashes and burns, if the interest in AI wanes even a little bit, you're going to see a serious pullback that I'm sure would lead to some kind of recession. So that would be potentially the big story of next year. But the.com era went from roughly 1995 to March of 2000, so there was a good five years there. If there is a pop, it might not be next year, it might not be the year after, it might be three years from now, who knows? But if it happens next year, that'll be the biggest story of the year.
- [25:40]
A
Anything else you want to add or.
- [25:41]
B
Just a final thought, this was the most fun year to cover tech in a long, long time because again, so much is up in the air and I get excited about that, even if there's negative things about it. Not knowing what's going to happen and not knowing what the next new thing is like, a deep seek is. It's. It's exciting for someone like me.
- [26:02]
A
Thank you so much to our guest Brian Mikola for joining us for this year in review. Definitely. Go check out the techbrew Ride Home daily podcast if you love all things tech. We will continue to cover some tech news right here on the Newsworthy's weekday roundups, but Brian always provides a whole bunch more. And by the way, he recently started a sort of hobby podcast he wants you to know about called Rad History. It's a history podcast, but only about the 80s and 90s. It's super fun. Again, it's called Rad History. We'll link to both of his shows in our episode notes on our website thenewsworthy.com just click episodes and find today's date. Thanks again for joining us today. As always, join us again during the week to catch our 10 minute daily news roundups. We use a fast, fair, fun style to keep you in the know on the latest news and headlines without getting bogged down by arguing pundits or clickbait and rabbit holes. We're back with more news on Monday. Until then, have a great rest of your week.