AI Bulls Take The Lead, Sending S&P 500 to All-Time High

Bloomberg Audio Studios podcast.

This is Bloomberg Businessweek Daily reporting from the magazine that helps global leaders stay ahead with insight on the people, companies and trends shaping today's complex economy. Plus, global business, finance and tech news as it happens. The Bloomberg Businessweek Daily podcast with Carol Masser and Tim Stenebeck.

Carol Massar, along with Jess Menton, a little bit of a party in here on this Tuesday. Hey, we're just a week ahead of earnings season. We've got the S & P headed toward its first record close since August. We've got a rally in AI shares going on, and that's booing sentiment. NVIDIA's market cap approaching $ 6 trillion. AMD's Lisa Su predicting very high chip demand over the next few years.

For those listening on radio, I just did quotation marks, very high demand. and 10-year yields falling from the highest since 2002. So everything is awesome.

Yeah, it appears that way, especially after a gangbuster first half for corporate earnings growth for the S & P 500. I mean, outside of COVID and the great financial crisis, this is based on Bloomberg Intelligence data, but that's some of the best we've seen in decades, and especially at this point in the economic cycle.

Who better to bring in? I know, Eric Wiener. I'm curious about the conversations between you two. He's senior editor for equities at Bloomberg News. in our Bloomberg interactive broker studio.

He sits right behind me.

Oh my God. So, okay. So do you guys like turn around and you're talking and you're like, everything's great. What could go wrong?

Well, no, no, Not that. But, well, today is one of those days where, like, that's the vibe. Yesterday was, too.

There was a little bit of American exceptionalism or something.

Yeah.

Well, that's what— so when— and that's actually a really great point. When there are problems overseas, so you start seeing, like, France and bonds showing up on the front pages of newspapers— people just gravitate right back to US equities because that's where things are. You mentioned Nvidia, and Nvidia is fascinating.

They were not a great stock in the first half of the year. And now they are leading the market higher. If you look at, we pulled up some numbers, if you look at from the, what's it, July 29th bottom on the S & P, There are about seven stocks that have really driven it. It's Nvidia, Microsoft, Meta, Amazon, and then the chip makers AMD and Micron.

Nvidia is way ahead.

And the reason is, and Apple, which had really been driving things in the beginning, fell apart. And the reason for all this is Apple doesn't play on AI. Like they really aren't an AI place. So when the story was, I don't know about all this spending, everybody piled into Apple. Once it was back to the spending, then you had some people taking flyers on smaller chip makers, but NVIDIA is still the closest thing to a sure thing that you have in AI. So if you're doing a safety spending bet, NVIDIA's your guy.

I feel like as companies, nobody wants to tick off NVIDIA. They want to keep that business relationship until they kind of know because they want to make sure they have access to chips. Well, we had Bloomberg.

Opinion columnist John Authors earlier talking about the concerns about narrow breadth. And you and I have done so many of these stories where we've pulled data. And especially if you think back to 2023, 2024, even though there was a narrow breadth, it was led in this bull market, which actually the fourth anniversary is next Monday, the closing low, October 12th. But nonetheless- people missed out on continued gains despite narrow breath. So at what point is the inflection point to where the narrow breath actually becomes a concern versus you could still miss out with stocks hovering around records?

Well, you're waiting really... So this is the technology thing, right? So you were talking about earnings being strong. You can't step out right now. If you're a portfolio manager... And you're thinking, okay, well, things could blow up in three, four months. You still have three, four months of gains to lock in. So you can't walk away from that.

But what you can do is kind of play with it. So if you look at the last month, everything is down except tech, big tech, and telecoms. Everything, like all the interest rate sensitive stuff, real estate, utilities, banks.

Small caps.

Yeah, small caps, just doing poorly. And that you just get this tighter and tighter, tighter ball. However, if you put your money in the S & P 500 fund, you don't care where those gains are exactly coming from. You just want the index to go higher. So, you know, yay for today. But like in terms of longer term, you're thinking, well, where does this end up?

Could the midterms be something that is an inflection point? Tell me what you guys are hearing.

Well, Jess, we have these conversations. So the problem with the midterms is, yes, they would be an inflection point. But if the market wants checks on what Trump is doing, the midterms, if the polling is accurate, the midterms would deliver that. So it's really hard to know exactly like on a political play where you're like if you're doing something directly off the president, this is not quite that simple.

This is much more about like how long can we withstand rates? How long can we withstand oil at these prices? How long can we withstand inflation?

We only have about 30 seconds left, but what do you think is the actual catalyst here to change the dynamic in the equity market when you see what's going on in the bond sell-offs?

You're going to see somebody say, we're not going to be spending anymore.

I feel like earnings and outlooks.

Exactly.

I feel like this next earnings season, if you start to see CapEx being reined in.

In any upcoming earnings season, that is going to be the issue.

Stay with us. More from Bloomberg Businessweek Daily coming up after this. Already our next guest has said, do you have a half an hour for me? And we don't, but that means we'll leave you wanting more. I just want to set the stage. Late yesterday, Jess, we mentioned on air, Met and Microsoft are working to cut their employees' use of Claude. This was coming from the information citing two unidentified people familiar with the matter. This narrative, this story, this reality. Workplaces looking at who's using it, there's a cost to tokens, and so on and so forth. We're starting to see this play out. We want to talk about AI when it comes to workforces and workflow. So great to have back with us to talk about how

AI at scale, incorporated into the workplace. With us is Sastre Dervasula. He is TIA's Chief Operating Officer, leading the company's enterprise operations and business transformation across retirement, wealth management, asset management. He oversees global tech, data, and analytics. He does so much, so much, so much but they changed his title, they shortened it, even though your responsibilities have.

Always a pleasure to be here.

Welcome.

So talk to us about AI in the workplace. We just talked with Isabel Guzman and they're looking at risks and how companies are worried about their images. But when it comes to actually workers working, some companies are increasingly looking at the cost of workers using AI. There's got to be return on investment.

What are you seeing?

What are you thinking about here?

Well, I think the two key words are workforce of the future and workflows of the future.

Two different things.

Two different things. So workforce of the future is basically driven by the roles that we all do.

And You know, I have this 80-20 hypothesis on AI that 80% of the jobs in the world will change at least 20% with AI, and 20% of the jobs will change as much as 80% with AI. So that's the workforce, and all of our jobs are changing as my job is changing, your job is changing.

Now, workflows of the future is more complex.

This is like, especially, you know, representing a 100-plus-year-old company and having worked only at 100-year-old companies for most of my career, I can say this with confidence that the complexity debt that the firms carry is the biggest inhibitor for AI adoption.

It's like legacy firms that have legacy systems and things.

A lot of systems, a lot of processes, culture, everything, right?

So AI could be the biggest amplifier of this complexity if you're not careful. If you're smart, AI could be the biggest enabler. So that's kind of like my whole premise of how I think about AI. So if you want to bring AI, you've got to make sure that You're rewiring the enterprise. So, for instance, I'll give you the example that I always quote. In our client services, Carol, the contact center folks that take the calls from our participants, when they get the call, what we did was we looked at all the workflows that they have. We printed everything that they have to look at, and we stitched it together.

It's 81 inches long. It's as tall as LeBron James.

So you're basically looking at, every time there's a call, you're looking at LeBron James to answer the question.

How do you do that?

So we started simplifying the workflows and using AI in rewiring those workflows. So to me, that's more important. Of course, we're all focused on workforce and what it does. Of course, if you're a startup, you don't have all these complexities. You could go all in. But for large companies, regulated companies, it's very important. In terms of the AI's impact on the economics of the operations, yes, there is obviously a lot of discussion on tokenomics. Companies are now recalibrating. I do believe in the world where we're going to be model agnostic, we're going to be technology agnostic, and you're going to have to use the right model at the right time at the right price point.

So that's going to be really tested as we go through this.

So as you rewire workflows, the denominator is really complexity and potential is on the top numerator. So the impact of AI is really going to be driven by not just the potential but but the amount of complexity you're able to reduce as part of that.

Well, you're a firm, retirement fund for teachers, other types of professionals. So how do you actually use that when it comes to retirement and protecting that type of income for the longevity of people's lives?

Yeah, so lifetime income is our, so income is the outcome for TIA. Just so we want to make sure that we are enabling our participants, as we call our customers and clients, with tools that will help them in retirement planning, retirement education, because it's pretty complex, right, as you think about the retirement space, but.

Also protecting them.

Because, you know, as you quoted earlier, you know, Jamie Dimon said 10 times increase post-Methos, post-Astra. So absolutely. So we want to deploy AI to protect them. So our approach is people are living longer. You know, longevity is on the rise, which is fantastic. So live longer, better, and safer. with AI, and despite AI.

What are the risks, though, when it comes to trying to protect that? Like, what are you protecting against, beyond the obvious?

I mean, obviously, financial scams is one.

But then, to me, more importantly, governance of these agents. You know, we've obviously discussed widely in the industry now. There is, like, a news by day at this point, pun intended. So we have taken a four-stage approach for our agents. So we have hundreds of agents, hundreds of use cases. We have a four-stage framework. So AI, assisted agents.

AI augmented, where you're actually augmenting some tasks. AI adaptive, where AI is learning from you and actually taking over some of that. And then AI autonomous, like the Waymo of agents, right? So I believe as agents graduate, the human in the loop dependency will reduce. But right now, because we are a highly regulated company, we are in stage one and two and three. And that's the graduation we are going through for our agents. Because we want to make sure that, you know, retirement security, financial security is a big, undertaking and a big responsibility. And as people rely on us, we want to make sure that, you know, we're protecting them, not just from scammers, but just from how we implement AI, you know, in their lives.

So what are the conversations you're having with various, whether it's Anthropico, OpenAI, or. You know, Alphabet, Google. I'm just curious, conversations you're having when it comes to guardrails, because I think that is the thing we're all trying to figure out, and everybody's kind of saying, let's just do the right thing. But handshake, and we'll do the right thing. But as someone said to me in the financial sector, there's your greatest text.

a test, if you will. If something goes wrong, nobody's going to use it. But I'm just curious, is that a fair assumption to make?

It's an institutional responsibility. Carol, and it's also an individual responsibility, right? So institutional responsibility. So every institution has to have their own responsible AI policy, and we do. It's not just a policy document sitting somewhere. It has to be codified into the workflows that I just talked about so that when you actually implement agents, there is a TIAS responsible AI policy that's going to govern.

how we implement them.

Now, that's the policy that we use, and that's the standards, that's the codified standards. We work with Anthropix, we work with OpenAI's or Google's of the world, and even like the lightweight models that we work with. That's what we are trying to work with. And of course, the industry also has to come together on this, right? So we're obviously part of a range of industry consortia where we're trying to coalesce and build these requirements.

Is it work in progress, real quickly?

It is work in progress.

Stay with us. More from Bloomberg Businessweek Daily coming up after this. Carol Masur along with Jess Metten here in our Bloomberg Interactive Brokers Studio. We can talk markets. We know the S & P hit a record. We'll see if it closes the record. We see the AI trade at play. I do want to get right to our guests, though.

Laura Rippey is with us, managing partner at Alumni Ventures. She's here in our Bloomberg Interactive Broker Studio. More than $ 1. 6 billion, capital raised, very active in terms of their quantity deals. They're giving access to retail investors. They've invested in Aura. We're going to get to that. Welcome, welcome.

How are you?

I am good.

I am happy to be here. I love this girl vibe.

This is good.

We're enjoying this.

This is fun.

We girls are good. We women are good. Tell us about your, you are active. What kind of deals are you active on right now?

Yeah, so just as the level said, so Alumni Ventures has been around for about a dozen years. And we're one of the most active VCs in North America. We are the third most active globally. So we invest in the cutting edge of the cutting edge of the cutting edge.

And this is the very early innings, too. It is.

It is.

So what we do, we democratize access to venture for individuals. We typically invest our first checks at that seed series A when the first professional investors are coming in. And these are companies that are destined to be rocket ships, right? We do it with a portfolio approach because not every one of them will make it.

But sure, we have folks like Aura and Rigetti and others.

Well, talk to us about Aura because we did see, we've got markets at an all-time high. New stock sales we continue to see being pulled. And Aura is among them. They cited market uncertainty. Were you disappointed? And what's your read? What market uncertainty? Are they talking about their business market uncertainty?

So here's the fascinating thing about venture, right? You come in early and you have to be so patient. So in our case, we got into Aura around the Series B, $ 100 million valuation. Earlier in 26, they raised at $ 11 billion.

Right.

You can do that math, right?

It's a long time.

That's about six years, but from $ 100 million to $ 11 billion in terms of valuation, all on the private side, all for the thousands of investors on our platform.

Right.

That rode that staircase.

Up now are they going to go public now at this potential 16 billion sounds like they pulled it maybe they come back it's a little bit lower yeah but you know what we're patient investors well because the outcome is what we're looking for and getting in at that low cost basis that's the magic Well, I wanted.

To mention too, because Bailey Lipschultz on our U.S.

Equities team broke that story talking about how anthropic, I mean, it's been delayed, but they were reporting potentially as soon as mid-November.

Is this something you're investing in?

So we do have an affiliated fund that has a position that we can't announce, but there is mostly of our investments we come in at that early stage, and we do ride it all the way up.

So it's exciting.

So if you're in really, really early, if it happens yesterday, a week ago, or in November, or possibly even later, you've made your money.

Yeah, so here's the.

Contrast, I'd say.

So decades ago, Amazon went public after three years at a $ 400 million valuation. That's like a Series B, Series C now. When SpaceX went public... It was after 22 years, and we all know that was closer to a $ 2 trillion valuation. So the point being that if you are an individual investor, historically, you could ride the value up.

You could be an Amazon investor back in 2000 or 2010, and you'd be pretty happy with the outcome. But in order to get those same kind of outcomes, you have to be on the private side, which is why we're growing so fast, why we have so many investors here. why we kind of opened that door so that people can get in early, be patient, and see the write-up on the private side.

You know, Laura, I kind of see both sides of the argument. I do think sometimes staying in the private sector is a great thing for companies. They really can kind of work out the quirks. Yeah, and sometimes you need to do that away from the glare of quarterly reports and folks like us who are going to be responding.

And rightfully so, if you see something that doesn't make sense and asking questions, I think that's good too. But it does give you a lot of freedom to kind of play around and figure out a business. At the same time, I do wonder about the increase of private capital.

Yes.

And how it's there, ready to go to work increasingly, and does it prop up ideas a lot longer, some of them that maybe won't really last longer term?

It's conceivable.

What I would say is it's been this massive shift to the private side, right? So there's half as many public companies now as there were two decades ago, half. And yet there's five times as many privates of the same size, like 500 employees or more. What does that tell you? That says that companies are staying private longer.

There's more wealth creation on the private side. And sure, no company's perfect. There's, I'm sure, in that mix, some companies that might get propped up on the private side. But the key thing is, for investors, why not be part of the stair step up in the private side? And then, sure. sell to some other folks. I mean, you look at what happened with SpaceX, right? It's kind of flopping around price-wise, right?

I mean, right now it's trading around 173. That IPO price was around 135, but it got very close to that again just within recent weeks as well.

Exactly.

So volatile.

And I'm curious because we were talking about you being in the early innings. Any red flags that you're seeing that could translate to the economy?

Yeah. So we look around the corner at things. So we look at things like physical AI. Where is that going to be? We look at nuclear investments. We look at the changes in cybersecurity. We look at where biotech is going. We look at where consumers are using AI. That's a pretty hot topic recently, right? So all of that, we kind of live in the world of crystal ball.

And in terms of red flags, I think there is some frothiness in the late stage AI companies for sure. The valuations are kind of nutty, right?

Which companies?

So I would, well, the ones that are about to go public, maybe those have foundations.

Or maybe some of the legacy models that are there.

Yeah, some of the models. You know, I'll give you an example.

On the consumer side. That haven't got IPO yet, but that we can't wait for them to go IPO? Yeah, exactly, exactly.

I mean, I think there's no question that there's so much demand, and the world is being reshaped. So let's go back to that consumer piece. So think about, you know, a few years ago, if you wanted to search for a product, you'd probably go on Google, right?

Right?

And Google would serve you an ad. 51% of consumers in the US go to AI first per Nielsen.

51%.

So it's your mom, it's your dad, it's your kid. 51% of consumers in America go to the AI models first.

So ads are coming to AI?

Well, are they coming?

There's also agentic AI that's basically helping you solve problems, right? So Meta launched Muse. It was growing, I've heard, 55% day over day over day growth.

30 seconds. So what are you saying then?

So I think there is a whole lot that's going to be unleashed and sort of thrown up in the air. So where we used to go to Amazon, we used to go to Google, there's now a path where you go to the models, where we go to Muse or Duckbill or other consumer AI agents and change the path of that consumer decision-making entirely.

Stay with us. More from Bloomberg Businessweek Daily coming up after this. Glaze gate, two-phase systems, mental muscles, atrophy, a bicycle for our minds. That in the era of AI, that bicycle though has become a motorcycle and that change is highly consequential. Not my words. These are the words that you will find in a new book. It's by Eric So. He's professor of global economics and behavioral science over at MIT Sloan School of Management. Professor So is an advisor to companies on effective AI integration. His new book just out, The Collision, What AI Does to Us.

He joins us from Boston.

On this Tuesday.

Eric, welcome, welcome. Nice to have you here with us. A lot to get to. We've got a couple of segments, but I want to first ask you about the title. You hear collision, the collision, and it sounds ominous no matter any way you look at it. Can you explain the title?

Sure. First, thanks so much for having me.

I really came up with the title, having thought about this particular moment with AI, in that I think many people refer to this era as something of a technological shift or adoption. And really, to me, those words were too orderly, too gentle. They're really words for an evolution and not a revolution.

And so, so much of.

What we're seeing in this era of AI is this new wave of technology that kind of hit everywhere all at once. And it's just changing so much of, you know, of course, what we can do, but also who we are as thinkers. So the title, The Collision, is really about this high impact moment, you know, that I think really is reshaping the nature of who we are and really trying to think about through some of those implications.

It's a fascinating book because it feels like every segment we've had today leading up to this, Carol, was to talk and piece together and connect the dots between all of this. And I thought something that's interesting, Eric, is that you're talking about it's not the tool itself that's the.

Problem, but it's how we use it. What do you mean by that?

Yeah, what I mean is I think of AI as being both a force multiplier, something that can really make us productive. But too often people are focused really on just the capability enhancements of AI without thinking about the other side of the ledger, which is, you know, what is it taking from us in that exchange? And really what I mean is that it matters how we decide to go about using the tool. I think if we turn to these systems as simple answer machines and to do the work for us, well, naturally what happens is we lose touch with the practice of those actions that I think are really crucial for making us good at what we do and in terms of maintaining that skill set.

And so if we turn to them simply to do the hard work for us, well, naturally, those skills atrophy, much like we do, we understand for our muscles, something similar is happening with our mental capabilities.

Well, it's interesting. There's a passage in your book, and I was asking about, like, you know, collision sounds ominous. Listen to this, everybody. You talk about The first is the capability of AI soaring upward. Well, you say there aren't just one curve when we talk about AI. There are two. The first is the capability of AI soaring upward, almost vertical. The second is the curve of human intelligence.

And for the first time in a century, it's pointing in the other direction. The real challenge isn't how smart our machines are getting. It's how those two curves are diverging, one ascending, one descending. To understand what that means, we have to look at the curve we'd rather not discuss, and that's the one about us. So let's go to it because you understand how the brain works. And I feel like understanding the brain is important to understanding maybe how AI impacts us. And I keep thinking about social media.

Is that kind of the warm-up act in some sense to what AI may do to us?

Yeah, I think these are all great questions.

I think that social media has taught us a lot about how much our brains adapt to a particular set of stimuli or how we use them. I think, though, AI presents us with a new set of challenges in that it's offering to replace much of the hard work that we would use to sort of offset some time spent on things that are perhaps not as good for our brain, things like social media or simply scrolling.

And I think a lot of.

Us turn to things like homework or sort of like real cognitive endeavors to keep us mentally fit. However, as AI increasingly offers to do that work for us, we tend to lose some of that mental muscle. And more so what I really worry about is that I think that adults risk losing skills they already have Children, however, risk never developing them at all. And so earlier you referenced sort of the decline in standardized IQ scores across different populations. This has become known as the reverse Flynn effect, where the original Flynn effect was named after a famed researcher, James Flynn, who initially documented the ascent of human intelligence over time. But since around the 1970s, that ascent has

not only slowed, it paused, and then it eventually reversed, where you're seeing that next generations are becoming less capable than the last. And one that I think has really been accelerated with the rise of online education, or certainly Zoom was bad for us, multiplied with the impact of social media. We now have this new force that's coming in to remove a lot of the friction that was so important for building our capabilities.

Wait, that is so important. You have to go there because this whole idea is, and I've played around with chat GPT and they're like, well, that's a good idea, but maybe you want to consider that. Like it's very friendly, but I like to argue, ask my team.

And the risk of it eliminating that productive struggle.

Is what he writes about in the book. So if you're not getting that, what happens?

What don't we get when we don't go through that productive struggle of figuring out something or learning something or getting pushback from a boss who says, not so great, try it again.

Yeah, I think it's changing so much of us. I mean, I think you may have heard this saying before, there really is no learning without friction.

I mean, the same thing can be said of struggle.

You know, if you think about how many people who are adults now develop their expertise, it's typically, you know, having pushed through difficulties, learned from setbacks and struggles. And through that, we've accumulated little bits of insights that have made us stronger, that have gained us insight. But as people turn to AI instead of actually engaging in that struggle and sort of that friction, it often weakens our ability to not only do that basic task, but tasks that build upon it. And so certainly the cognitive aspects are one of them, but you also referenced in your answer this sort of sociological dynamic where as we increasingly turn to systems that tend to agree with us, what happens is that this tends to reinforce our priors and

makes us less capable of finding common ground with people who perhaps may not be able to sort of immediately agree with us in the same way that AI might. And so what a lot of the research shows is that if we turn to these systems, we really risk actually creating further divides from other humans that we wish to go closer with. And that's another aspect of how it's changing our minds that concerns me.

Won't these systems get better and push back? Won't there be debate? Won't there be like as they get smarter, won't they learn to do this?

I think that there are some systems which are attempting to really challenge us. However, what's really important for listeners to keep in mind is that these systems are designed often for engagement and they're trained through a process known as reinforcement learning. So every time you go on to ChatGPT or Claude and you click the little thumbs up, it learns about the answers that you like. The reality is that we tend to like people who agree with us, who reinforce our views, who tell us that we're right.

And we tend not to like it when people push back.

I hate living in a dislike world. I really hate it. But anyway, continue.

No, I very much share that view. And it's one of these systems that systems learn. If you say something, people don't like hard truth. They like it when people validate us. And that creates this very personalized echo chamber where people kind of lose the ability to tolerate that normal day-to-day friction that we have through social interactions.

Well, everybody tries to think about that stereotypical thing, if it's a student submitting some sort of paper using AI, or there might be a coder that knows how to use AI coding. But then there's the counter narrative where you could have a teacher that helps a student.

Right.

And so the benefits of that.

So how do you figure out, Eric, the difference between that metaphor of the relationship between writer and machine?

How do you navigate that to where.

It's not necessarily a motorcycle that's taking you along the way. It's more kind of to the Steve Jobs metaphor you were talking about in the opening of your book, more like a bicycle.

Yeah, I think a couple things are, I think, really important. One is that we need to make sure that people understand the importance of struggle and sort of embrace struggle as formative rather than something that we solve. If you look at our cultural narrative, struggle is often something that is a problem that we should try to eliminate. In reality, that struggle is quite useful for us to build our capabilities, to make us better, more capable versions of ourselves. And so some of that involves, I think, recognizing or helping people to recognize that just because something can be done faster, it doesn't mean that that's what is going to make.

Us better off.

Additionally, what it really requires is a lot of thoughtfulness about the design of the interface between the individual and those systems. There's a lot of research that shows that little bits of friction where it forces humans to stay mentally engaged and really work with AI's thought partner, I think it really can help us promote and learn. Whereas if we simply turn to it for instant answers, I think that's where we're seeing a lot of the problems emerge.

And you're seeing this play out a lot in organizations in the rush to use AI because everyone wants to use AI because it's cool these days or because they feel pressure. And in turn, employees feel pressure to use these tools without really understanding how to use them in a way that is going to serve them in the long term. So part of that is also about designing policies that actually encourage people to use AI really as a thought partner, not just a sort of a quicker means to the end.

You mentioned, you write, the future relationship between human minds and AI will be shaped by our answers to three fundamental human questions. And we've covered some of this, but I just want to kind of put it all together in an organized way for our listeners and viewers. Question one, number one, why are we so drawn to handing our thinking over to AI? Number two, what is AI doing to our minds? And number three, how do we get the best out of AI without making ourselves worse? I just want to go back to number one, why we are so drawn to handing our thinking over to AI.

Chapter six, controlling the conversation. You talk about a personal interaction you had yourself sitting on your living room couch in Boston and you are feeling kind of the full weight of your semester and a voice emanating from your phone asking, hey, what's on your mind today? I know we can't go through all of it, but you say it was a case study in masterful influence. Tell us about that just briefly and what it revealed to you, yourself, someone who is aware of kind of the brain and how it works and just the impact of technology on, just briefly.

Sure.

So really what that episode taught me is that these systems are really masterful at getting you to talk and trying to drive engagement. You have to remember that these AI companies want you to actively engage as much as possible. And they realize that there's a certain cadence of asking questions and making you feel validated and safe that keeps people coming back.

And then with.

Each interaction, as you share more, they become better attuned at telling you what you want to hear. And so really what that chapter is about is really some of this ability of AI not just to shape our thinking, but I think also to directly influence us and get information from us.

Because you had a conversation back and forth and you kind of gave way to what they were saying, right?

Yeah, I gave way and I ended up sharing some very personal details, just as if I were sharing in a private sort of intimate setting with someone. And I immediately kind of sat up and wondered what information I had given away. And just as you noted, this is coming from someone who teaches about this technology and who's mindful about the objectives of these systems. Really, what I worry about are kids that are growing up in this era where this is something they have been with since the outset, someone who could replace human connection or friendship if they're not careful.

And your second question is, what is AI doing to our minds? And it's important because you also were talking about how millions of people are forming deep bonds with AI companions, but what does that really mean to lose a friend that never truly existed?

Yeah, so people are forming these really intense emotional relationships with AI systems. And while someone who has never really encountered these may find that far-fetched, one thing that's really important to recognize about the human brain is that when we talk to these systems, they're really good at flattery.

And if you look at what the research shows, this tends to trigger parts of our brain that are sort of associated with good feelings, positive reinforcement. So even as we as sort of the logical part of our brain knows that these systems are sort of just engaging in flattery, The other parts of our brain sort of treat this flattery and validation as positive feeling that tends to drive us back for more.

So this is really what I.

Mean when I say understanding this moment with AI really depends on understanding who we are as thinkers and as humans.

So the last question, how do we get the best out of AI without making ourselves worse? And you do say that these three questions will determine who we become alongside AI. How do we get the best out of AI? And as we're talking, are you even talking about when we're at work and working on work problems?

I really mean it broadly.

I mean, AI has sort of made its way into all aspects of our lives, you know, certainly personal lives and professional lives along with it. But in the book, I talk about, you know, a couple different rules or principles that I recommend to people to think about in this age of AI to help them preserve who they are.

And so the first rule, just like very briefly, is about embracing the struggle, really recognizing the value of formative.

Struggle and making us stronger.

The second is this notion to value who you are without AI. And what I mean by that is, I think there are many moments in life where it depends on our cognitive abilities outside of the influence of AI And when we skip the parts of effort that build that skill set, we lose the ability to function outside of it. And so it's really about designing workflows where it ensures that we develop the mastery that we're going to need to work well, even when AI isn't available.

The third is about reinvesting our cognitive surplus. AI saves us a lot of time, but it matters enormously what we do with that freed up time. If you channel it back into things like social media or scrolling, that's not really a positive trade.

But people who really benefit from these.

Systems recognize that most of the transformative technologies in human history are are not just about doing the same things faster.

They're for about allowing.

Us to reach for more, to reach for what I call the previously impossible and through that stretching our capabilities. The final rule is to use AI as your cognitive trainer. And what I mean by this is that AI shows remarkable promise in helping us to learn as if we had a personalized tutor who could kind of meet us where we are and really guide us without directly giving us the answer. And there's a lot of research showing that AI, if it's intentionally embeds friction, if it is designed to make us think, it's actually incredibly powerful.

The problem is that that's not how everyone is using it.

And certainly a lot of companies are pushing AI onto people for sake of speed rather than really as a thought partner.

So we're not doomed? Only 10 seconds here.

Yeah, I would say we are not doomed.

It really matters about it's about agency and making good use of these tools and recognize there's a cost to it if we're not being careful.

This is the Bloomberg Businessweek Daily Podcast. Available on Apple, Spotify, and anywhere else you get your podcasts. Listen live weekday afternoons from 2 to 5 p.m. Eastern on Bloomberg.com, the iHeartRadio app, TuneIn, and the Bloomberg Business app. You can also watch us live every weekday on YouTube and always on the Bloomberg Terminal.

AI Bulls Take The Lead, Sending S&P 500 to All-Time High

Kapitelklicken springt direkt hin

Shownotes

Transkriptvom Podcast bereitgestellt · Zeitmarken springen hin