New Era Of Tech Accountability? + The Impending AI Ethics War

Ranjan Roy of Margins joins us for another Big Technology Podcast: Friday Edition covering the week's news. This week, we check in on the following stories: 1) JPMorgan's failed acquisition of Frank, a college financial planning platform with an exaggerated userbase 2) SBF's new Substack 3) The SEC vs. crypto 4) AI's impact on the tech giants 5) The impending 'AI ethics' war 6) Is anyone still using Mastodon? For weekly updates on the show, sign up for the pod newsletter on LinkedIn: https://www.linkedin.com/newsletters/6901970121829801984/ Learn more about your ad choices. Visit megaphone.fm/adchoices

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Alex Kantrowitz?

  This episode is brought to you by Genesis. What does it actually take to put Agentic AI to work across an entire enterprise? I recently spoke with Genesis chairman and CEO Tony Bates about why orchestration is emerging as a competitive advantage in customer experience, and what companies need to get right in the AI era. Then I sat down with Adam Mitchell, head of enterprise business solutions at Voya Financial, to go inside a large-scale Agentic CX transformation, including the business decisions, trade-offs, and lessons along the way. You can watch both conversations now on my YouTube channel today. It's never been more challenging to get frontier compute. How do you know where to look, who to trust, or how much to pay? Akash is an open cloud that provides easy access,

• 0:45Alex Kantrowitz?

  flexibility, and peace of mind. That's how companies like Razer were able to cut their inference cost by 15 times. Visit akash.network/alex for free consultation. That's akash.network/alex for free consultation.

• 1:17Alex Kantrowitz?

  Welcome to Big Technology Podcast Friday Edition, a show for cool-headed, nuanced conversation of the tech world and beyond. And we're here once again to break down the day's news, the week's news, as we will do every Friday. I'm joined, as always, by Ron John Roy of Margins. Ron John, welcome.

• 1:34Speaker 01

  Thanks for having me back here.

• 1:36Alex Kantrowitz?

  It's great to have you back. So we're going to start this week's show. We have so much to cover. We're going to talk about accountability. We're going to talk about AI ethics and the fraught nature of what that field might bring, and maybe the battlefield that we're going to see with that field. And then we're going to talk a little bit about some of the big stories with AI, the competitiveness with the tech giants, and of course, Microsoft's investment, or potentially pending investment in OpenAI. But last week, we ended on a fun story. This week, let's start on a fun story. There's this company, Frank. I'd never heard of Frank. Had you heard of Frank?

• 2:11Speaker 01

  I had not heard of Frank, but I also have not been a student in a long time.

• 2:15Alex Kantrowitz?

  Exactly. Okay, so Frank is a company that sold to JP Morgan for $175 million. Essentially, it was lead gen for JP Morgan to reach younger customers, students, and it exaggerated its customer base, at least according to JP Morgan's lawsuit that it's bringing. By an order of magnitude of maybe 10, it really had a few hundred thousand customers, but it thought it had four million, and the lawsuit has some unbelievable facts inside of it that Matt Levine really did a great job breaking down this week. The overarching theme, though, here is that we're in an era of accountability

• 2:58Alex Kantrowitz?

  that potentially we weren't in previously, where like in the past, it might have been a situation where it's overlooked or whatever, and now it's front and center. And this person, the founder of this company that sold it to JP Morgan, is now getting sued. So what do you think about this? What do you think about the story? What do you think about it in terms of what it heralds?

• 3:22Speaker 01

  Yeah, when we say that the numbers were exaggerated by a factor of 10, I think that's understating just how crazy this story is. So for a bit more background, the platform Frank, they said they had 4.3 million customers when they were sold to JP Morgan. JP Morgan had pressed them during the due diligence to provide them more information about this. At first, the founder had resisted under the guise of data privacy and user privacy. But then still, the JP Morgan pushed, so they had to come up with a list of 4.3 million users. So what they did, they actually went out, first they tried to have someone internally do this.

• 4:07Speaker 01

  That person ended up leaving the firm because they refused to. Then they went to a data scientist, and actually, we'll probably be talking about generative AI and ChatGPT later. They were early adopters on this because they actually worked with a professor and using, basically to create synthetic data to take their dataset of 300,000 actual users and create an entire new world of 3.1 million users. And that's what they submitted to JP Morgan to actually sell for 175 million.

• 4:41Alex Kantrowitz?

  It's a pretty freaking crazy story.

• 4:44Speaker 01

  Yeah.

• 4:44Alex Kantrowitz?

  There's so many questions to be asked here. How did this even get past JP Morgan's due diligence department? I mean, how did JP Morgan not, not look at Frank's email addresses beforehand? If that's what was really something they were interested in.

• 4:57Speaker 01

  I mean, obviously, my first answer will always be Zerp because this still happened in the

• 5:03Alex Kantrowitz?

  Oh, it did happen during the Zerp, Zerp days.

• 5:04Speaker 01

  Yeah, this happened, I believe it was 2019.

• 5:08Alex Kantrowitz?

  Right.

• 5:08Speaker 01

  But even then, it's still, it's, it's such a reminder that when an acquirer fits a very specific profile, again, JP Morgan trying to do things that are more innovative, trying to reach a younger user base, trying to reach the next generation of financial consumers, clearly will get a bit of ahead of itself and probably let certain things slide. Also, they did have an outsourced consultant that apparently did vet this and prove this during the diligence. So

• 5:40Alex Kantrowitz?

  Maybe they should sue them.

• 5:42Speaker 01

  I mean, I guess in terms of intent from who really tried to defraud who, it's pretty clear from everything that's

• 5:50Alex Kantrowitz?

  It was Frank.

• 5:50Speaker 01

  come out here. Yeah. And, and I think it's, the idea that they thought they could get away with it, and they did, is actually the most fascinating part to me, that something that's so over the top and clearly fraudulent, the fact that a founder still went ahead and did this and tried to get away with it when clearly at some point you will be caught at some point. And I mean, in the, the story of how they actually were caught, JP Morgan did try to email 400,000 of the clients of the emails on the list. I think it was about three quarters of them bounced. Almost there was an incredibly low click rate, open rate. And from there, that's when they started to dig a little deeper.

• 6:34Alex Kantrowitz?

  And the real question here is, are we going to start to enter this, this age of accountability, which I teased at the top here. There's a great story about this. First of all, the journalism on this has been unbelievable. It is always wild when A, a company buys something that really isn't very real, or allegedly not very real for so much money, $175 million. That's crazy. But then typically, and this is something that Eric Newcomer brought out in his newsletter. Typically what happens is founders will just, or investors, acquirers will just distance themselves from the fraudulent asset. Think about, you know, a Quibi, not fraudulent, but not really real. Think about Quibi, right? It wasn't like, you know, the investors-

• 7:14Speaker 01

  I did not think Quibi would be coming up on this Friday, but I'm glad to hear the name.

• 7:18Alex Kantrowitz?

  But like when stuff fails really poorly, or it doesn't really hit the way that investors think, usually the investors just kind of shirk away. But in this case, JP Morgan is suing. And this is from Eric Newcomer. He says, "These kind of lawsuits are so rare. I have to believe there are many cases where a public acquirer simply writes down their investment and tries to prevent shareholders from figuring out that the acquisition was an obvious mistake from the beginning. I wonder," and this is the key point, "if founder accountability might get a jumpstart this year as the downturn in private tech continues to play out." That's a really interesting thought. I hadn't thought about it that way. And maybe Eric is on to something here.

• 7:56Speaker 01

  Yeah, I will admit, it's not off the top of my head. I did just Google it. But JP Morgan had $78 billion in free cash flow last year. So the idea that they would just normally write off some investment like this where clearly things didn't work out, clearly they were fleeced, I think in the last few years, or five years ago, would have been the case. I do completely agree that right now, I think there is a complete mindset shift where-

• 8:25Speaker 01

  Businesses realize that they have to start holding this type of behavior accountable. And it's because that this was allowed to go on for so long that this is happening. And again, I think it's clear from everything that's coming out, just how obvious this fraud was. And I think that's why they chose to go after it. But I think we're seeing that in a lot of different arenas in technology now where things that were allowed to be slipped under the rug. Clearly Theranos was the big story a few years ago, but there wasn't much else. I think we're going to see between this, we're going to see more and more instances where if it's just complete fraud, you're going to, I mean, they're going to go after you.

• 9:04Alex Kantrowitz?

  So the ages, I mean, one of the interesting things that people are bringing up with this Frank story, again, so many accounts that weren't real, then getting effectively packaged into a product that JP Morgan bought was, didn't they know that this day was going to come eventually? I mean, is it indicative of the days that we've come from and to the days that we're going to that this type of stuff was allowed to happen? It's pretty interesting to me. What do you think?

• 9:28Speaker 01

  Yeah, I, I think that, so, so in this case, from what I, what I saw in the coverage, she had actually sued JP Morgan two days before they filed their lawsuit. So it appeared to be a bit preemptive. So, and clearly the more that came out, there was a lot of bad blood that she had come in as the head of student solutions and very quickly things went downhill. Once they started uncovering all of these fake email addresses and it became essentially a head to head battle. So I think, I think in this case, again, there appears to be a bit of that back and forth where they may have just written it off if this did not get personal. But at a certain point, when the fraudster tries to come after you, once they actually

• 10:13Speaker 01

  fight back in some way, I mean, JP Morgan is not going to take this lying down. And, and my, my favorite part of this is in her complaint, the founder, Charlie Javis, about JP Morgan, first of all, her lawyer is Alex Spiro, who is

• 10:29Alex Kantrowitz?

  Elon Musk's lawyer.

• 10:30Speaker 01

  Was or is Elon Musk's lawyer heavily involved in the Twitter, in the entire Twitter drama was reportedly a big part of the transition team when Elon took over. And let's remember. He was on the other side of actually filing the lawsuit saying that Twitter was having fake users last year. So I mean, clearly a litigator who is happy to play both sides. But, but it, when you see all this coming up.

• 10:57Alex Kantrowitz?

  Good point. I hadn't even thought about that. Yeah.

• 10:59Speaker 01

  Yeah, I know. I mean, yeah, Alex Spiro, I feel there's going to be more that comes out that's interesting around just him as a character and all of this, everything that's going on in tech right now. But, but, uh, yeah, I think the fact that JP Morgan is going after this publicly and aggressively, I mean, it's a huge sign that founders will be held accountable. And, and I mean, again, SBF, Gemini today, all of these other areas, we're seeing people being held to account.

• 11:29Alex Kantrowitz?

  So let's move to those couple of stories. First of all, SBF and Substack. SBF is obviously not keeping quiet about the allegations against him. He spoke to one of my friends, Teddy Schleifer at PUP News. Teddy wrote a terrific story about it. And he also started a Substack. I mean, come on.

• 11:46Speaker 01

  I can't even.

• 11:47Alex Kantrowitz?

  That was astonishing. So I, full disclosure, I didn't read it. I understand that he has a story to tell. I've heard enough about it at this point. And I didn't personally want to hear any more about it from him. I found it interesting. Okay, he's doing a Substack. It was funny. He initially was charging people or like had a paid option. And then he's like, oh, that was a mistake. He turned it off. What, what do you make of the fact that, I mean, is he's in this position? And he probably, I don't know, maybe he should be talking, maybe he shouldn't be talking. It's, it's interesting. The traditional corporate playbook is if you're in a place of

• 12:22Alex Kantrowitz?

  being prosecuted, don't talk. But he has this view that he wants to talk and talk more. It's very interesting.

• 12:29Speaker 01

  Well, full disclosure, I did read it. I tried to trudge and power through it. And it was the same thing over and over. And again, you can try to argue that.

• 12:42Speaker 01

  I guess the most generous interpretation I'll give is he has been consistent. He has with a straight face told George Stephanopoulos, told, I can't even remember who, Andrew Ross Sorkin, whoever else, that I did not steal money. This was just a bet gone wrong. This was a leveraged exchange gone wrong. That happens, the London Metal Exchange a few years ago, like over levered exchanges can go bankrupt. Hedge funds can make bad bets. He just keeps saying it over and over. He keeps saying that there were letters of intent to fund FTX up to the tune of $4 billion, right up to the bankruptcy. He keeps saying these same things over and over, even though everything has shown that they're just completely untrue.

• 13:26Speaker 01

  I mean, Caroline Ellison, Gary Wang, Nishad Singh, everyone has been coming out. And it's clear in all the filings that are coming out that this was just fraud from the beginning. This was funneling money from FTX to Alameda and personally making political donations, buying luxury real estate, having, what is it, like $100,000 catering bills? None of it was, you know, just a bet gone wrong. I'm sorry, this happens. This was just outright fraud in terms of actually, you know, misrepresenting to customers, saying your account is safe with FTX when you're just funneling the money over to Alameda. And he's still just not answering that simple fact.

• 14:09Alex Kantrowitz?

  Okay, so the investor, Bill Ackman, I want to get your take on this. Had a thread about this and he said he previously had been accused of something and found out that, and it came out that he wasn't actually guilty of it. And we know that Caroline Ellison is guilty because she's pleaded guilty. We don't know about Sam. When you hear that, what do you think?

• 14:30Speaker 01

  So the Ackman thread, I did read. The one bit of empathy, sympathy, I might feel for him is I do think the more I learned after the fact, I remember living through it, but being younger, Elliot Spitzer, and in terms of kind of over, over zealous for political purposes, regulation, I do think, you know, he definitely pushed the envelope in the other direction. And so when you see that name, maybe Bill Ackman was the target of something that might have been a bit over zealous. However, still comparing that era of, you know, like hyper regulation, really

• 15:12Speaker 01

  aggressive policing of Wall Street, Henry Blodgett being barred from the industry, you know, like whomever else, there was every number of cases that were going on. I mean, trying to compare that to today, this is what we're talking about. There has been no accountability over the last few years. So when regulators are going after you, when the SEC, when there's criminal complaints, I mean, right now, I think the, it's very clear that there's a reason. The envelope is not pushed so far in the other direction where they're going after absolutely everyone. I mean, you are the biggest marquee fraudster of the moment. You're going to be a target.

• 15:52Alex Kantrowitz?

  Now, speaking of targets, let's just do one more crypto story and then we can move on to some of the other stuff that we want to talk about. The Winklevoss twins, by the way, starting, I didn't realize, but it has become common lingo that people just call them the Winklevi, which is hilarious. Like I heard them called the Winklevi on CNBC and I was like, oh my God, that like meme from the social network is now reality. They are in a bit of an issue. The SEC is coming after them for their earn program. And it's this convoluted relationship between them and this company called Genesis, where people were effectively, imagine they loan their crypto out, they get eight percent on the money. I mean, you have to ask, what, why am I getting this eight percent? I think that's

• 16:34Alex Kantrowitz?

  one of the lessons we're learning with these crypto firms. And ultimately, they haven't been able to get the money out. And now the SEC is coming after them, saying this was a security. What is your read on what's happening here? And is this sort of, are we at the end here of these type of things with crypto? Or is this still, are we still in the middle? I guess we haven't had Binance yet. I was talking about that with Kate Rooney from CNBC, who was on earlier this week. But what's your read on John?

• 16:59Speaker 01

  I mean, we're certainly in the middle because Sam Bankman Fried in that Teddy Schleifer piece is still living a pretty comfortable existence in a large Palo Alto home. The Winklevoss are still tweeting. Like, I mean, we're definitely still well in the middle of whatever is going to happen. I do think maybe we're approaching the end game because this is the point where everyone is pointing the fingers at each other. Everyone, you know, the good vibes of good morning, we're going to make it are long, long gone. And now it's everyone else is a crook. And the Gemini story, I think it's very interesting because as you said, Gemini,

• 17:42Speaker 01

  and it gets a bit confusing because there's Gemini and Genesis.

• 17:45Alex Kantrowitz?

  It's really an annoying story to follow.

• 17:47Speaker 01

  I know, even reading through every article and trying to track which one is which is tough. So Gemini, they had their earn program, promised exorbitant yield. To get that yield, they lent that money to Genesis, which is under the Digital Currency Group, Barry Silbert. And then Genesis has had some trouble from the FTX fallout. They're not able to give that money back. And this is where you're getting letters between Tyler Winklevoss and Barry Silbert and everyone saying whose money is who. And, and that is, I mean, that's the one, all of this is Sam Bankman Fried has $450 million of Robinhood shares. Does BlockFi get it in their bankruptcy?

• 18:32Speaker 01

  He's claiming that it's his and it's had nothing to do with FTX. There's all these pockets of money that's left and of fiat money or Robinhood shares, which theoretically are liquid. So, so everyone is fighting over them. And, and the biggest thing about today, the SEC or yesterday, once they filed this enforcement action against the Gemini. The thing in terms of accountability, this made my blood boil in terms of Tyler Winklevoss. He was tweeting and I copied this down, that this action does nothing to further our efforts and help earn users get their assets back. Earn is the program Gemini. Earn, their behavior is totally counterproductive. And he says, "Super lame. It's unfortunate that they're

• 19:17Speaker 01

  optimizing for political points instead of every..." This is exactly what SPF is doing. Somehow you lost everyone's money and now you're saying you are the one who should be entrusted to get back their money. And then the last one I noted in this thread of his, "We look forward to defending ourselves against this manufactured parking ticket." I mean, the condescension towards the largest regulatory body of the United States government, it's so palpable there, but it is. You have not had any pressure. You just lost theoretically $900 million of your customers' money and still they're just not treating this as a

• 20:01Speaker 01

  huge deal, as a criminal deal potentially. They're treating this as, "It's an issue. We're the ones who can get it back and you guys are screwing up our efforts to do well by our customers."

• 20:12Alex Kantrowitz?

  I mean, the way that they're talking, and I also copied down the part, the point where they were like, "Super lame." I mean, talking about the SEC, like it's residual from a moment where there wasn't accountability, but maybe what we're seeing right now is accountability is making a comeback, at least in some way. That's the hope. Do you think that that's the case?

• 20:33Speaker 01

  I think so. Matt Stoller, who of the antitrust fame-

• 20:38Alex Kantrowitz?

  Yeah, we gotta get Matt on with us on one of these Friday episodes.

• 20:41Speaker 01

  Yeah, that would be a fun one. One thing I really have taken away from things he's written is that government needs to govern and has forgotten how to govern. And this is something for me over the last few years, I really think the whole attitude of deregulation, I mean, and obviously this extends to everything, airlines, we're seeing issues right now, whatever else, but especially around financial markets. This attitude of like, you see it again, super lame manufactured parking ticket. They're calling out the regulator who has just filed an enforcement action against you. And like making fun of them essentially because they're so used to not having any type of being held to any type of accountability.

• 21:25Speaker 01

  And I do think, I think again, SBF, if he goes to jail with the Winkleville, who is responsible financially, does more stuff come out? We'll wait to see with Binance. And that's all in this area. But again, going back to Frank and JP Morgan, the fact that their founder is still aggressively coming out against JP Morgan and saying that there's something around like JP Morgan just wasn't ready to help us attracting a young, diverse, new audience, that they weren't ready to do that. Like just saying ridiculous things when you are under the microscope and like caught red handed around doing very bad things.

• 22:07Speaker 01

  I think once we see more action, this, I mean, this era of accountability will definitely be here to stay.

• 22:14Alex Kantrowitz?

  We're here on a new Friday edition of Big Technology Podcast. Ron John Roy is with us. We started doing these last week. It's a new thing we're doing in 2023. The flagship episode on Wednesday, where I interview newsmakers, talking deeply about what's going on in their world is still going to continue. And we're also going to do these shorter ones on Friday where we break down the week's news, which is what we've been doing in the first half. If you like it, please rate us five stars on your podcast app of choice. And if you really like it, please stay tuned because we'll be back right after this break. This episode is brought to you by App Point. Everyone's racing to roll out AI right now. Co-pilots, chatbots, agents doing real work. But here's the part nobody loves talking about.

• 22:56Alex Kantrowitz?

  All that AI runs on your data and most teams have no single way to see it, secure it and prove it's under control. That's exactly what App Point does. For 25 years, they've been the trusted layer beneath the world's most demanding data, now extended across your entire AI estate, your data, your

• 23:16Alex Kantrowitz?

  Cloud and the agents acting on your behalf. It's how more than 28,000 organizations deploy AI with confidence. So innovation scales without scaling risk. AVPoint, the unifying trust layer for AI. Head to AVPoint.com to see how enterprises deploy AI with confidence. Learn more at AVPT.co/bigtechnologypodcast. That's AVPT.co/bigtechnologypodcast. Violent threats against executives are growing at an alarming rate. When a company experiences backlash, executives are the first people blamed. When their address and other personal details are sitting online, this backlash can lead right to their front door. The team at Ironwell knows this better than anyone.

• 24:01Alex Kantrowitz?

  They've protected some of the most targeted executives and individuals on the planet for almost two decades. As a wide glove enterprise security service, they protect your people with continuous personal data removal, proactive prevention tools and human-led emergency support. So when someone goes looking for your executives, Ironwell ensures they hit a dead end. So here's what to do. Go to ironwell.com/bigtechnology, fill in the quick form and request your free risk assessment. Their team will show you just how exposed your executives are and how to lock it down before a threat reaches their front door. That's ironwell.com/bigtechnology. Learn how to stop online threats before they become real world attacks. The link's in the show notes.

• 24:46Alex Kantrowitz?

  And we're back on Big Technology Podcast Friday edition. Ron John Roy is here with us, second week in a row. Great to have you, Ron John.

• 24:57Alex Kantrowitz?

  Oh, you're muted.

• 25:00Alex Kantrowitz?

  It's Friday again. Let's do this. Let's do this. So by the way, we're going to do these live on LinkedIn every single week, but I messed up the technology this week. So we're not live, but I will post the video. Not this week. No, we got, we got some questions last week. We did get questions, so next week I'll make sure to get it done. Join us will be live 11 a.m. Pacific, 2 p.m. Eastern, 8 p.m. Central European Time, if you roll that way like I am at the moment. And let's get into the second half. So obviously, artificial intelligence is a huge. Topic of interest for anyone focusing in on the technology world right now.

• 25:39Alex Kantrowitz?

  There was a really excellent piece by Ben Thompson in Stratechery talking about how AI is going to impact the five big players. And they're obviously interesting things about apps on Apple being powered with new technology that can lead to things like image generation with an app like Lensa. Thing like Meta using AI to do better targeting of ads now that Apple's cut off its ability to target precisely.

• 26:04Alex Kantrowitz?

  One of the things that I found interesting reading through this article is Ben thinks this is a real shift change, a shift change in the way. Getting the PC was going to the internet was moving to cloud was moving to mobile was and now. There's this new shift and of course it feels big as we're watching it happen. If you've used chat GPT. If you use Dolly, if you use anything of this nature, you realize that we're in a new world right now. But my question is, how much is it going to shake up the business world? Because we did, we did have Aaron Levy from Box on a couple of weeks ago, who really thinks that this is a sea change. And I know that you read Ben's piece, Ron John. Did you think that it put the changes that we're going to see with this wave of AI into the appropriate context?

• 26:46Alex Kantrowitz?

  And it's also interesting that this was all happening all along the way and right now we're talking about it because we have these generative use cases, but it's more than just the generative AI. So what's your read on on the magnitude of the shift here?

• 27:00Speaker 01

  Yeah, this is one area where, I mean, Ben's piece I think is fantastic at really laying out the concrete use cases for where artificial intelligence is going to change these businesses. As you said, each company, each of the big five, Microsoft, Meta, Alphabet.

• 27:21Speaker 01

  Everyone has their own, Amazon has their own specific use cases. And I do think this is going to completely change the way companies operate, business is done. The Meta, again, they actually Business Insider just had a really good long piece around. It was like Meta's advertising business is back. After the Apple tracking efforts in iOS fourteen, I think it's now a year and a half ago, and losing the tracking abilities in their, like advertisers and agencies are starting to say that it feels like it's back and that has been very

• 27:56Alex Kantrowitz?

  I was waiting for that story. Yeah.

• 27:58Speaker 01

  Yeah.

• 27:59Alex Kantrowitz?

  It was inevitably, it's going to tick back a little bit and then inevitably going to be a story where someone writes, oh, everything's hunky dory again because advertisers are moving a little bit of their spend there. But I think that's overblown, but go ahead.

• 28:11Speaker 01

  You think? All right. All right. No, I think it, but theoretically, like, let's take a meta, like, just kind of like, heavy-handed tracking, like cookie-based tracking, app or device-based tracking, that is going away, that we're moving away from that world. So instead, leveraging artificial AI to actually create models where still being able to create user profiles and still be able to target in new innovative ways, it seems theoretically possible. And if anyone is positioned to do it, I think they are. Obviously, I think the one everyone is excited about and interested in is Google search versus, I mean, if Bing becomes a name and a player

• 28:56Speaker 01

  in the 2020s, that would not have been on my bingo card a year ago. But the, I mean, this idea that Microsoft, and we can definitely get into their relationship with OpenAI, I mean, Satya Nadella is the GOAT CEO right now because carefully fostering this relationship and having firsthand access to GPT-3, OpenAI, all their technology. And again, chat, we talked about this last week, search will change. Search already has changed. Google search is a mess right now, but they've already started, Google is the famous one with snippets, trying to, without you going out to another page, try to extract what is the relevant answer to your

• 29:40Speaker 01

  query and trying to give it to you. I mean, large language models create an entirely new way of doing this. I still am skeptical about Casey Newton on your show saying that he will find a new pair of shoes via ChatGPT, but I do think, I mean, this will change search. It's inevitable. If anyone who's used the technology. Being able to synthesize a large number of websites and then into a coherent, simple answer. I mean, that's what search, the promise of it always was, and it's going to change a lot of the internet because the advertising-based internet, click here, use Google to go to my website so I can run an ad against it. I think that's going to change and everyone needs to start preparing for it.

• 30:24Alex Kantrowitz?

  Right, and Ben seemed most, to be most pessimistic about the future of Google. But let me now stand on the side of Google and talk about why Google is actually going to be better. Of course, it's a more, that the old Google search might be better, just for the sake of argument. Of course, it is going to be fun to talk to these things. But there is definitely a utility in being able to poke around different websites and do your own research. Because whether we understand it or not, part of our way of interacting on the internet is looking at different sources and making determination of what we believe after we triangulate. That's, I think, an important part of using the internet, and that's going on for everybody.

• 31:06Alex Kantrowitz?

  And if you have a chatbot that is going to get things wrong, and inevitably will, then you're going to end up, or it will present one perspective. Because I can sit there and talk you through all these different perspectives for every question you ask it. So I do think that it's going to be, it's going to be tougher than a lot of people imagine for that type of experience to replace the classic search. I think classic search is pretty evolved.

• 31:30Speaker 01

  However, on the other side, generative AI, I do think it's like most potentially malicious impact, call it, is going to be just millions and billions of websites and content created trying to game SEO. And completely, Google needs to change the entire way to approach its search in order to deal with the coming influx of just mass-generated content. They've said in the past, like Danny Sullivan, their search liaison, "We will penalize AI-generated content." That was before ChatGPT was public, that was before, I mean, we will, like, having used, played with this technology a lot.

• 32:10Speaker 01

  I mean, the idea that there's going to be very simple ways, or even complex ways of being able to perfectly tell what is AI-generated and what is not. I don't think is going to be feasible. And I actually think there's going to be a lot of good AI-generated content, and that should not be penalized in any kind of search. So I think Google is going to have to rethink the simple idea because it was built on a foundation of a website that is on the internet has inherent value. Someone put effort into it, and now we are going to rank these. And obviously, there's always been black hat SEO and just, you know, content farm-y things. But but I think the scale of that is going to change massively.

• 32:52Speaker 01

  So yeah, I do. I do think from a search perspective, this creates challenges on both sides for Google. On one side, you have this idea that now your search query is returned to you in this highly personalized narrative form. And then on the other, the actual way Google search works, I think is going to be impacted.

• 33:12Alex Kantrowitz?

  Well, you you use the one argument that that hurts the most for me and sort of backed me into a corner because I've been the victim of this this week.

• 33:21Speaker 01

  I read your piece. Tell us about it.

• 33:23Alex Kantrowitz?

  I have a story up on Big Technology about a publication on Substack that was started up a week ago. And on Saturday, published a story that was a not a carbon copy, but a plagiarized copy of the story that I wrote about the creator economy the week prior, using exact sentences, but also having remixed a lot of the content and spitting out this new article. And the interesting thing about it was the writer who put it together, who was anonymous, called themselves Petra, ended up admitting in the comments after this after this plagiarized post went to the front page of Hacker News, which is worth thousands of views to any publication. This person admitted in the comments that they had used AI to remix or for to quote unquote,

• 34:08Alex Kantrowitz?

  improve readability. And actually, they meant, you know, they never cited the fact that the original content was coming from me. So on that point, yeah, I think that there's some serious, there's a serious issue with it. I just would, I also wouldn't want ChatGPT to be like, oh, you want to learn about the creator economy? You know, here's a summary of what Alex just wrote, and then not having anybody come to my page. If that happens, I'm in trouble. And so is the entire, I mean, those type of applications really hurt the whole web. But it's gonna be.

• 34:40Speaker 01

  It it it's changing everything. I I I really think, I mean, from our as writers, we have to think about this as Google, how they rank webpages, they have to think about it.

• 34:54Alex Kantrowitz?

  Right. And no one's gonna wait, raise their hand and be like, oh, by the way, I used AI to write this. And Google, please de-rank me. That's the same thing with these. Substack wouldn't take the wouldn't take the post down. The the AI generators had no way to find the person. So the traceability of all this, because at the end of the day, it's just text, is is really difficult.

• 35:12Speaker 01

  No, no, and

• 35:13Alex Kantrowitz?

  We're gonna have to live with it.

• 35:14Speaker 01

  Who is to say that a person did not take your post and just rewrite it?

• 35:18Alex Kantrowitz?

  Exactly.

• 35:19Speaker 01

  I mean, plagiarism has happened since since the beginning of time. So yeah, no, I think it is really important and that there needs to be like open, honest conversations around this right now because it is going to change how everyone approaches just content in general. And I do think like, I mean, the Sam Altman, whoever else, they need to be leading these discussions. And I think the ethical considerations around this are huge. And and I'm actually incredibly bullish on all this technology because as someone who has written very boring content projects, I can tell you, there's a lot of the world and there's a lot of, there's a lot of copy produced in the world

• 36:03Speaker 01

  that is not exciting for someone to produce. And if the more that gets automated, great. However, what are the other implications of this plagiarism, someone rewriting your post? I mean, this stuff, I agree that there needs to be, we need to at least start thinking about what the guardrails look like.

• 36:21Alex Kantrowitz?

  Did you see that CNET is now having AI write posts for it? Straight up, write articles for CNET.

• 36:28Speaker 01

  Yeah, I I mean, so I I thought so. I had looked at, I've been like following the space very closely. Even years ago, there was, I think it was like 2015, there was a company called Narrative Science. And they, I think it was Reuters or the Associated Press. Yeah. And again,

• 36:43Alex Kantrowitz?

  Forbes also used them. Yeah.

• 36:44Speaker 01

  Okay, yeah, but but and it made sense. Earnings reports, sports scores, take it basically, but that was basically essentially a fill in the blank type of uh, type of service. It's like we have these essentially what are templates. Earnings report comes out, give the headline, there's some if then statements around if, you know, like the earnings per share are lower than the estimate, you frame it in a more negative way. That stuff's been around for a long time. And in some ways, you could argue that, you know, ChatGPT is essentially a smarter version of that. But, but yeah, I do think, I think journalism and content providers will be incorporating more and more of this technology.

• 37:26Speaker 01

  And I think we just have to figure out what that means. I'm still a little skeptical, though. Whenever anyone says, there was definitely a while where like, you know, this article was written by GPT and like, no, you, you spit out a draft and then you edited the entire thing, but.

• 37:40Alex Kantrowitz?

  Yeah. Okay. Conclusion of this is I do think Satya Nadella made some very smart bets with his closeness to OpenAI. And I'm about to speak with Todd Bishop about it. Todd is the founder and editor in chief of Geekwire. And it's the like, Seattle trade publication that covers the tech out there. And I'm very eager to hear what, what his take is on, on the whole Microsoft situation, especially when it comes to OpenAI. So listeners, if you're interested, check that out this upcoming Wednesday, Todd Bishop coming on, onto the podcast. But Raja and I agree with you. The Satya thing is, he's obviously the GOAT CEO at this point.

• 38:14Speaker 01

  Yeah. So to put it into context, one of the most brilliant maneuvers, I mean, when I actually, I remember when Microsoft invested a billion dollars in OpenAI in July 2019, everyone's interested, this is where, where are they going with this? Apparently what happened was a good amount of that was Azure credits. We touched on this last week. It's brilliant because ChatGPT is the greatest marketing tool that has ever touched the space of language, large, large language models. I mean, apparently, Google has a highly functional comparable.

• 38:51Alex Kantrowitz?

  Well, that's Lambda.

• 38:53Speaker 01

  Yeah. And I mean, they invented the transformer model, which this is all based on. So they should be pretty far ahead. But instead, suddenly, every, making ChatGPT essentially free, exposing everyone to this and branding yourself around it. That has been an incredibly, incredibly expensive thing to do. But instead, what happens with Microsoft, the brilliance of this investment, they invest the credits to OpenAI. OpenAI uses those credits to build its brand. I mean, it's almost like. Okay, this will be a stretch, but like a crypto firm giving tokens to uh to its service as the investment dollars. Yes, but

• 39:34Alex Kantrowitz?

  Web three was real.

• 39:35Speaker 01

  This was web three. It's Microsoft gave OpenAI tokens. They use those tokens smartly. And that's the reason everyone is talking about chat GPT and not any other com- open source service. Google's still private, quiet with whatever it is doing. So I think that that move alone was brilliant.

• 39:55Alex Kantrowitz?

  You know that every cloud conference that's gonna happen for the next two years is gonna have a big Azure advertisement that says, use the tools that chat behind chat GPT or something like that.

• 40:06Speaker 01

  Yeah, yeah, yeah. And apparently, apparently, uh, OpenAI before that investment was already spending one hundred twenty million a year with Google Cloud. And then because of that investment, yeah, it was in the information, uh, because of that investment, switched completely over to Azure. And now, I mean, from a branding standpoint, just overall, again, Satya Goat CEO.

• 40:28Alex Kantrowitz?

  So for our final topic, we we don't have a lot of time, but and we could probably spend a full show on this. Maybe we will in the future. But let's talk a little bit about AI ethics, because there is gonna be a battle that goes on, uh, pretty blatantly about what happens with the ethics of these models. And ethics in some ways is, of course, it's a good natured way to make sure that they're not causing harm. But the undertone or the undercurrent of all these conversations is that people want to control what the models do. And there's gonna be a battle between, just like there has been in content moderation, there's gonna be a battle about what ethics should these AI uh, models have. It's not just like, should they be ethical? It's what ethics should they have? And there's this

• 41:10Alex Kantrowitz?

  post that Sam Lessin put together about how AI ethics is the most dangerous part of twenty twenty-three AI. And he talks about, he talks about, uh, he says, the more, the morality limits that companies like OpenAI place on GPT, he says, I find the morality limits that companies like OpenAI place on GPT three so deeply troubling. It's a free speech problem, but a massive steroids where a small group of people are taking a technology slash a fundamental power and then centralizing the mortality morality power of what it will and will not allow the tech to do. This is just a preview of what the big debate is going to be. I think content moderation debate is going away. I think the AI ethics debate is coming.

• 41:56Alex Kantrowitz?

  And I'll just say what I think about it, and I'm curious to hear what you think. But my perspective on this is that it is a little bit naive. Because when you create a model like this, you ultimately are making decisions about the content, just like when you create an algorithm and you build in functions like the retweet and the like, you are making decisions about what to elevate or not. So you do have an influence on content. And the fact that you are the company that's making the product means that you are a small group of people that are influencing it. So this idea that we shouldn't put limits, or we shouldn't have a discussion about it, or there should be no content moderation, there should be no AI ethics.

• 42:36Alex Kantrowitz?

  I mean, obviously, it's naive. There's no should or should not. It exists. And that's the debate that's going to happen. So I'm curious what you think about it, Ron, John.

• 42:45Speaker 01

  Yeah, Sam, first of all, Sam, I think has perfected the art of like fitting a blog post into a tweet with a screenshot.

• 42:51Alex Kantrowitz?

  I wouldn't call it perfected. I hate reading those things.

• 42:53Speaker 01

  You hate reading? I like it. I like it. Whenever I come across one of those.

• 42:57Alex Kantrowitz?

  Yeah, go ahead.

• 42:58Speaker 01

  I'm a fan. It's the Apple Notes extension. But, okay.

• 43:04Speaker 01

  I, first of all, I'm so happy that this is a conversation right now, because if you think about it, you know, like, six years ago, this wouldn't have even been to, uh, this wouldn't have even been anywhere on anyone's radar. We would have just built, built, built, and then gone and whatever happened, we would try to clean up the mess after. So I think it is a good thing that, you know, like, there's a lot of smart, responsible people who are tackling this. The other thing is, I'm actually not that concerned, because ChatGPT, sometimes it could feel a little arbitrary as to what is blocked, what's not. Dolly definitely sometimes feels like,

• 43:46Speaker 01

  you know, you try to type something and you're like, this isn't that bad. And it's just like, sorry, we cannot produce this image. However, I don't think, this is like a bigger view of where this whole world is going. I don't think these fairly generic large language models are the end solution and the end goal, and that's how we will interact with them. I think already OpenAI, the business case and the value is going to be other models like DaVinci, Babbage, Curieita, that you can actually fine tune and train for your own use cases. That's where businesses are going to go. That's who's going to use them. And again, I think there's going to be all types of fine tuned models. There's probably going to be a white supremacist faced fine

• 44:31Speaker 01

  tuned model. There's going to be, if Sam's worried about woke speech, he doesn't have to interact with that type of model. I think like it's going to get narrowed down for specific use cases and we won't, we will move away from this idea that there's going to be just one overarching model that controls all speech. Because I do think right now that is what ChatGPT is for most people interacting. And I do think it's crazy that I don't know who at OpenAI is making these decisions. I don't know how they approached it. Clearly, they're concerned about the topic of content moderation or societal harm. And I guess that's a good thing, but being done behind closed doors, I don't think is

• 45:16Speaker 01

  good either. So, so I, but, but I don't, I think in the medium term, that's not as much of an issue. I think, I think all of this stuff is going to become much, much more customized, fine tuned. And I think we all types of models will be available, services will be available.

• 45:34Alex Kantrowitz?

  But just because it's going to be distributed, I don't think we've seen the end, or really the beginning of this AI ethics fight. There, there will definitely, because the people that control the technology will inevitably set the ethics or, or be held accountable for the ethics. And that's going to be really interesting.

• 45:49Speaker 01

  I don't know. Maybe the government will be involved and the government will govern somehow. Is that terrifying that will sound to?

• 45:56Alex Kantrowitz?

  But do you really want the government coming in and setting rules for what AI can and can't do?

• 46:01Speaker 01

  I mean, do I want Elon Musk governing what can and cannot be tweeted? Is that a better solution?

• 46:09Alex Kantrowitz?

  Well, I mean, his governance is basically tweet whatever. Let's just end on this. Last, last thing I'll say, just let's think about Twitter for a moment. I mean, there was all this hubbub about it. I tweeted earlier that, that Twitter right now really kind of feels exactly the same as it was before Elon, except with bigger losses than it had previously.

• 46:30Speaker 01

  We made it till the end before

• 46:32Alex Kantrowitz?

  Right.

• 46:32Speaker 01

  The name Elon Musk came up.

• 46:34Alex Kantrowitz?

  Right, right. Well, you dropped it.

• 46:35Speaker 01

  Yeah, I know.

• 46:37Alex Kantrowitz?

  Hey, I've muted the, I've muted Elon Musk. I've muted Elon. I've muted Musk. It's made the experience a lot better. Not that I don't care what he's doing, but like at a certain point Twitter just became Elon Central. That being said, what do you think about the thesis? I mean, it doesn't really seem that that different. And I know you have a Mastodon URL in your, in your handle. How long is that going to stay?

• 46:58Speaker 01

  I, I log on to Mastodon a number of times a day. I think like I've limited, I still am on Twitter a fair amount. I will say I tweet less. Mastodon has been getting more and more interesting for me as a way of finding information, interacting with some of the people that I would interact with on with Twitter before. But I do think at this exact moment, um, I don't think, yeah, I can't think of anything crazy or main character-ish that Elon's done for a couple of weeks now. And whatever the case of that is, maybe the Q4 was so bad that even he realized, maybe I should tone it down. Maybe Tesla's stock dropping, however many percent kind of forced his hand on this.

• 47:42Speaker 01

  But, but at, at this exact moment, from a user standpoint, I think everything might be quiet. However, I don't know if you saw Tweetbot, apparently

• 47:53Alex Kantrowitz?

  Oh, yeah.

• 47:54Speaker 01

  The API was cut off. There's glitches. I mean, if he cuts off Tweetbot, if they start making aggressive API decisions like that, I think he'll be right back in the center of conversation and, uh, and I don't, I don't even know who, who, what were the most recent scandals? There's been plenty. They'll fade away and he'll be right back.

• 48:15Alex Kantrowitz?

  That's right. Well, I think that will bring us home for this edition, this Friday edition of Big Technology Podcast. It's been fun, Ranjan. Thanks for coming on.

• 48:25Speaker 01

  Thank you.

• 48:26Alex Kantrowitz?

  We can do a whole episode next week about the future of Elon versus Mastodon. But I'm sure we'll have plenty to talk about. Maybe we get Sam on when we talk about AI ethics. That'd be sweet. All right, everybody. Thanks for listening. Great having you as always. Again, Todd Bishop from Geek Whalers coming on Wednesday and then Ranjan and I will be back next week.

• 48:54Speaker 02

  Now, more than ever, we need someone on our side. But Rob Whitman? Whitman's votes drove up costs, cut Medicaid, and risked 350,000 Virginians' health care to pay for tax breaks for himself and his donors. He's not on our side. Shannon Taylor will fight corruption, cut taxes, and bring down costs. This November, vote for Shannon Taylor. She's for us. Paid for by HMP, theHouseMajorityPact.com. Not authorized by any candidate or candidate's committee.

• 49:24Speaker 03

  In this life, you can't win.

• 49:26Speaker 04

  When the world gets you down...

• 49:27Speaker 03

  No!

• 49:28Speaker 04

  ...escape to Pluto TV.

• 49:29Speaker 03

  This is objectively fantastic.

• 49:31Speaker 04

  Relax into the movies and shows you already know and love, all for free.

• 49:35Speaker 03

  Schwing!

• 49:36Speaker 04

  Even our news is uplifting.

• 49:37Speaker 01

  You stay classy, San Diego.

• 49:39Speaker 04

  Watch School of Rock, Crazy Ex Girlfriend, Anchorman: The Legend of Ron Burgundy, The X-Files, Wayne's World, and so much more.

• 49:44Speaker 03

  We're not worthy!

• 49:46Speaker 04

  The best things on TV really are free. You just have to know where to look. Escape to Pluto TV. Stream now, pay never.