Latest / The Chuck ToddCast / Interview Only w/ Dr. Ayanna Howard - Higher Ed In The Age Of AI

Transcript

Interview Only w/ Dr. Ayanna Howard - Higher Ed In The Age Of AI · The Chuck ToddCast

0:00
Speaker 1: This episode of the Chuck Todd cast is brought to
0:02
Speaker 1: you by Chapter. Now, almost everything you hear about Medicare
0:06
Speaker 1: is somebody quietly trying to steer you somewhere. And you're
0:11
Speaker 1: skeptical of that, as you should be. I get it.
0:14
Speaker 1: I mean, there's a federal lawsuit going on right now
0:17
Speaker 1: over exactly this. Companies that branded themselves as unbiased while
0:21
Speaker 1: allegedly pointing people toward whatever would pay them the most
0:25
Speaker 1: as some sort of broker. So if your default is
0:28
Speaker 1: to distrust everyone in this space, honestly, I get it.
0:31
Speaker 1: I'm a bit distrustful these days myself. That's the whole
0:34
Speaker 1: reason I want to tell you about Chapter. They're independent
0:37
Speaker 1: Medicare advisors whose entire model is that their advisors are
0:41
Speaker 1: not paid to push any one plan. They're just trying
0:45
Speaker 1: to help. Yes, there's compensation that goes with it. but
0:47
Speaker 1: it's across the board is what they're trying to do.
0:51
Speaker 1: At the end of the day, they're advocating for you.
0:52
Speaker 1: They're not the insurance companies themselves. They're an actual person
0:56
Speaker 1: who actually goes through your doctors, your prescriptions, your needs,
1:01
Speaker 1: and then compares them to every single plan available to
1:04
Speaker 1: find the best one that will work for you and
1:06
Speaker 1: hopefully save you a couple dollars too. And here's the
1:09
Speaker 1: part that sets chapter apart. They don't disappear once you've
1:14
Speaker 1: been in They will keep advocating for you. They're kind
1:17
Speaker 1: of like almost like a concierge service. You're just there.
1:19
Speaker 1: I need help. You're like customer service. You know, the plan,
1:22
Speaker 1: you know, help me navigate this again. So you ever
1:24
Speaker 1: get hit with a wrongful medical bill, they'll help you
1:28
Speaker 1: fight it. They want to stay an advocate for you
1:31
Speaker 1: even after you've done this. So look, I get it.
1:35
Speaker 1: I think there's all sorts of space. You see a
1:38
Speaker 1: lot of people are throwing themselves in this healthcare space.
1:41
Speaker 1: What I like about these chapter folks, I've spent time
1:44
Speaker 1: with them. This is a very transparent organization. It is
1:48
Speaker 1: Medicare and all of its affiliated stuff can be very confusing.
1:53
Speaker 1: The whole point of this is to make this easier
1:56
Speaker 1: and less confusing. OK, so look, take the skepticism you
2:00
Speaker 1: already have in all of this, point it at your
2:02
Speaker 1: own plan for 20 minutes. It's free. There's no pressure.
2:06
Speaker 1: And then call chapter today at 980-734-4200. Well, my guest
2:18
Speaker 1: today is Dr. Ayanna Howard. She's the president of Spelman College, which,
2:23
Speaker 1: of course, is one of America's most historic institutions of
2:27
Speaker 1: higher learning. She spent most of her professional life, though,
2:30
Speaker 1: thinking about artificial intelligence long before many of us were
2:35
Speaker 1: thinking about it every day. She spent a decade at
2:37
Speaker 1: the NASA Jet Propulsion Laboratory. So this conversation that we're
2:42
Speaker 1: going to have, and she's got a new book coming out.
2:45
Speaker 1: In fact, hopefully you can find it right now after
2:46
Speaker 1: this interview, Rebooting the Machines, because it's going to come
2:50
Speaker 1: out in November. And it gets at a topic that
2:55
Speaker 1: I've had with quite a few folks, whether it's in
2:58
Speaker 1: the AI space, which is who gets to write the rules,
3:02
Speaker 1: who gets to decide what history is. And as you know,
3:06
Speaker 1: I spend a lot of time bringing on authors here
3:09
Speaker 1: who are essentially providing us better details of what actually
3:13
Speaker 1: happened in the history of this country. And needless to say,
3:17
Speaker 1: if we let AI write our history, we are going
3:20
Speaker 1: to have literally a whitewashed set of historical findings in
3:27
Speaker 1: what we have. And I know it's something that Dr.
3:31
Speaker 1: Howard's been thinking a lot about, and she's got a
3:33
Speaker 1: unique perspective here, head of a higher education institution and
3:40
Speaker 1: intimate knowledge of how this technology works. Dr. Howard, welcome.
3:45
Speaker 2: Thank you. Thank you. I appreciate it. And I'm interested
3:49
Speaker 2: in how this is going to evolve.
3:51
Speaker 1: Well, good. I mean, I want to start with, look,
3:57
Speaker 1: the decision to sort of leave your area of expertise.
4:01
Speaker 1: You're at NASA, right? That's like, you know, for a scientist,
4:04
Speaker 1: no matter, you know, that's for an engineer, right? That's
4:08
Speaker 1: a peak place to be. And now you're at Spelman.
4:11
Speaker 1: which I'm glad for Spelman you're there. You want more
4:15
Speaker 1: presidents of higher ed to come from doing stuff, not
4:22
Speaker 1: always from other academia, right? There's sort of two types
4:26
Speaker 1: of university, well, three types of university presidents. Those that
4:28
Speaker 1: sort of grow up in a university, those that are
4:31
Speaker 1: sort of politicians who get plucked, right? And then, And
4:36
Speaker 1: then there's this third category of people who've had successful
4:39
Speaker 1: careers who come in. So first tell me about the
4:43
Speaker 1: decision to go into academia after being at NASA.
4:46
Speaker 2: Yeah, so I am a non-traditional academic, and I always
4:49
Speaker 2: say that. I loved my experience at NASA. I loved
4:53
Speaker 2: having the shirts that said, you know, I am a
4:56
Speaker 2: rocket scientist, or this is what a rocket scientist looks like.
4:58
Speaker 2: I was really geeked out around that. But when we
5:02
Speaker 2: had the second shuttle accident back in the day, There
5:06
Speaker 2: was a conversation around the value of NASA. What does
5:09
Speaker 2: it mean? What is the science exploration thing about? And
5:14
Speaker 2: people don't realize, but they froze funding for a lot
5:18
Speaker 2: of the projects. Now, back in the day, you didn't
5:20
Speaker 2: get furloughed, though. You would come to work and you
5:23
Speaker 2: literally could do nothing. For a scientist, that gives you
5:26
Speaker 2: a lot of time to explore about, okay, what's next?
5:29
Speaker 2: And everyone said, oh, this happens every decade or so.
5:32
Speaker 2: So I decided to see what's next at that time.
5:36
Speaker 2: I wanted to do robotics. I was a researcher. The
5:38
Speaker 2: only place you could really do robotics at the time
5:41
Speaker 2: was NASA. You can go to automotive or you could
5:44
Speaker 2: do academia. So I put my toe in.
5:49
Speaker 1: You know, it's interesting what you just said there. I
5:50
Speaker 1: have a friend of mine that I grew up with
5:52
Speaker 1: who's a nuclear engineer. And he's been in this public-private
5:57
Speaker 1: sector for a long time, working for a private sector
5:59
Speaker 1: who's got big government contracts. And he talks about that
6:03
Speaker 1: frustration of working for the government. There are times that
6:06
Speaker 1: you just have to sit there because maybe there's a
6:09
Speaker 1: government shutdown, or there's only a budget continuing resolution, and
6:15
Speaker 1: there's certain restrictions on what you can do with your research. Boy,
6:20
Speaker 1: just hearing you bring that up, I was just immediately
6:23
Speaker 1: triggered by him. And he would just talk about how
6:25
Speaker 1: frustrating that was. It's just he goes, I'm just an engineer.
6:27
Speaker 1: I want to do work and they won't let me.
6:30
Speaker 2: I want to do cool stuff. Yeah. And that was
6:33
Speaker 2: my first experience around that. So I was at that
6:37
Speaker 2: that era that, again, every decade or so. And I'm
6:41
Speaker 2: one that I can't be idle. I get in trouble
6:43
Speaker 2: when I'm idle. And so I have to make sure
6:47
Speaker 2: that I have work to do and have some creativity
6:50
Speaker 2: and exploration in what I do.
6:54
Speaker 1: I was been thinking about, as I was looking at
6:57
Speaker 1: your material and understanding your background and thinking about how
7:01
Speaker 1: are we going to regulate AI and what do we
7:04
Speaker 1: do with it? And there's a part of me that says, well,
7:08
Speaker 1: you know, how did we handle air travel?
7:11
Speaker 2: Right.
7:12
Speaker 1: And at first it was very, we, government wasn't involved,
7:14
Speaker 1: then it was involved, you know, and we decided we
7:20
Speaker 1: didn't have the regulatory institutions that we needed. We created FAA.
7:24
Speaker 1: In the military, we created an Air Force. I guess
7:27
Speaker 1: you could say NASA was in some ways an evolution
7:30
Speaker 1: to that a little bit. But I was thinking about
7:32
Speaker 1: NASA and wondering, do we need a new agency a
7:36
Speaker 1: la NASA that is focused on AI? We do.
7:39
Speaker 2: Actually, in my book, I talk about regulations and saying
7:42
Speaker 2: we need to create an FAA-like entity around AI. So
7:47
Speaker 2: one of the things that FAA does is that they
7:50
Speaker 2: have some rules of engagement. Everything from if you are
7:54
Speaker 2: harmed by an airline, i.e. you're sitting on the tarmac
7:57
Speaker 2: for four hours, there's some compensation to you. If AI
8:02
Speaker 2: does harm to you because it collected incorrect data or
8:05
Speaker 2: had a bad outcome, What is the payment or the
8:09
Speaker 2: compensation that's provided to us? People are trained. Pilots are trained.
8:14
Speaker 2: Flight controllers are trained. Everyone is trained at the airport.
8:20
Speaker 2: Here in AI, it's basically if you can open up
8:22
Speaker 2: a webpage and read something, there's your training. And now
8:26
Speaker 2: you can develop. And so thinking about that element of
8:30
Speaker 2: an FAA for AI, I think is required because AI can...
8:35
Speaker 2: really cause harms, but it also would have some value.
8:38
Speaker 2: I mean, transportation is so much easier that we have flights,
8:43
Speaker 2: but we don't have as many crashes as there was
8:47
Speaker 2: when FAA didn't exist.
8:49
Speaker 1: Well, I mean, look, air travel is not 100%. We
8:53
Speaker 1: know this. There's risk with air travel. It's one of
8:56
Speaker 1: my favorite things to say whenever I hear, well, something
8:58
Speaker 1: is you're 95 percent certain about something. I'm like, would
9:02
Speaker 1: you go on a plane that had a 95 percent
9:05
Speaker 1: chance of landing, which means a one in 20 chance
9:08
Speaker 1: of crashing? You know, would you get on that plane?
9:11
Speaker 1: And when you put it in those terms, a lot
9:12
Speaker 1: of people it's my way of reminding people, you know,
9:15
Speaker 1: probabilities look differently when you put yourself in the situation.
9:21
Speaker 1: It's true. On something like that. So what should be
9:26
Speaker 1: our risk tolerance for AI? Because risk tolerance is a
9:34
Speaker 1: funny thing. 95% for everybody else seems fine. 95% for
9:38
Speaker 1: me seems risky.
9:41
Speaker 2: So I think about correlating risk to the potential harm
9:45
Speaker 2: that it could cause. We're going to go back to
9:48
Speaker 2: FAA and airplanes. So, you know, airplanes could actually take off,
9:53
Speaker 2: fly and land by themselves. But when you think about
9:57
Speaker 2: it through that, the risk goes up. And so what
9:59
Speaker 2: do we have? We have human pilots that mitigate that
10:02
Speaker 2: risk so that it is at 99.9999%. It's the human
10:06
Speaker 2: in the loop. And so I truly believe, and I
10:08
Speaker 2: have a philosophy that for high risk healthcare, as an example,
10:13
Speaker 2: we should have a human in the loop that is
10:14
Speaker 2: second-guessing that is putting their own judgment, that is overseeing
10:19
Speaker 2: because AI will never be 99.99999% because we as people
10:24
Speaker 2: are so adaptive and we change all the time.
10:28
Speaker 1: Well, that's easy to say and everybody seems to agree
10:31
Speaker 1: to it. And yet we can't even, I mean, look
10:34
Speaker 1: what we're doing now. And I'm torn and you must
10:37
Speaker 1: be torn as somebody who's an engineer that pushes the limits,
10:41
Speaker 1: meaning you want to keep finding, you want to keep
10:45
Speaker 1: experimenting But do we want to use the human race
10:48
Speaker 1: as our guinea pigs?
10:50
Speaker 2: And I don't. So there has to be a balance.
10:53
Speaker 2: I don't want the, you know, nuclear, we have regulations.
10:57
Speaker 2: We didn't always have that. And we know the impact
10:59
Speaker 2: that that has had. I do have a fear that
11:03
Speaker 2: if we don't think about the regulations, we don't think
11:05
Speaker 2: about the guardrails, that we will have something happen. And
11:09
Speaker 2: then everyone will then come to the forefront and say, yeah,
11:12
Speaker 2: we need to have regulations. But who is going to
11:14
Speaker 2: be harmed when that happens? And so there's that element.
11:19
Speaker 2: But then The problem is that everyone, all of our countries,
11:23
Speaker 2: all of our neighboring areas are still iterating and trying
11:27
Speaker 2: to figure this out. So as a U.S., if we stop,
11:31
Speaker 2: then it means that we will be left behind. And
11:32
Speaker 2: so we're in this really weird, precarious situation where other
11:37
Speaker 2: people are developing without guardrails. We have no guardrails because
11:41
Speaker 2: we want to keep developing. But something might happen that
11:44
Speaker 2: is detrimental to us as humanity.
11:46
Speaker 1: Right. Let me try pharmaceuticals. We don't let something get
11:53
Speaker 1: out to market until the FDA says it's okay. But
11:57
Speaker 1: we let some stuff into market sometimes before the FDA approves.
12:04
Speaker 1: Is that a model, right? Because with pharmaceutical companies, they're
12:08
Speaker 1: still allowed to innovate. They're trying to find something. Some stuff,
12:13
Speaker 1: they have trials. There's a process. If it doesn't work,
12:16
Speaker 1: they don't bring it to market. Should an AI model have,
12:20
Speaker 1: should that be the framework of an AI model?
12:24
Speaker 2: I think the models should be associated with the outcomes.
12:28
Speaker 2: So for example, we have AI for healthcare. Why don't
12:32
Speaker 2: we have something like the FDA that's already regulating pharma
12:36
Speaker 2: for healthcare?
12:37
Speaker 1: Okay, what does AI do? So you're saying you want
12:40
Speaker 1: to split it by specialty, that there shouldn't be an
12:45
Speaker 1: AI board generally that if it's a healthcare model, let
12:49
Speaker 1: people that are healthcare experts.
12:52
Speaker 2: Correct. So some of these things are highly regulated already.
12:55
Speaker 2: So we know healthcare is highly regulated field. We know
12:58
Speaker 2: nuclear is a highly regulated field. We know transportation is
13:01
Speaker 2: actually not highly regulated, but is regulated. We have these bodies.
13:05
Speaker 2: And one of the nice things is not US-based, right?
13:07
Speaker 2: There are these national, international organizations that think about transportation
13:12
Speaker 2: and airfare and flights and all of these elements. So
13:15
Speaker 2: they have a system. Why is AI thought of this
13:18
Speaker 2: totally different entity that it needs to be separated? Because
13:22
Speaker 2: I think if I have AI for, you know, helping
13:24
Speaker 2: me make calls, I don't actually want that to be
13:27
Speaker 2: heavily regulated. Yeah, sure. There's some spam calls and all
13:30
Speaker 2: of that. I don't really think I care. But if
13:33
Speaker 2: it's impersonation, if it's doing mail fraud, for example, well,
13:37
Speaker 2: then we have regulations with respect to that.
13:40
Speaker 1: Well, don't we already have that? Isn't that what the
13:41
Speaker 1: FTC is? I mean, I look at the FTC and
13:44
Speaker 1: I'm like, do your job, FTC. You know, if we've
13:47
Speaker 1: got surveillance pricing and things like this, that feels like
13:51
Speaker 1: we already have the regulatory body for this.
13:55
Speaker 2: We do. But I think this is the problem. fundamentally
13:59
Speaker 2: is that people don't understand AI. So when I think
14:03
Speaker 2: about this, people don't understand it.
14:05
Speaker 1: When you say that, what does that mean? When you
14:07
Speaker 1: say they don't understand it? Let me give you an example.
14:11
Speaker 1: I view it all as a probability model. I think
14:15
Speaker 1: it is all predictive. I don't mean that in a–
14:18
Speaker 1: so I go about it that way. And that's how I–
14:22
Speaker 1: that's why I know it's not 100% accurate, but why
14:26
Speaker 1: I assume it's finding the best guess. And it probably
14:29
Speaker 1: is better at initial search than me skimming a search page.
14:35
Speaker 1: But I also know it's just predictive and probability. But
14:39
Speaker 1: what is that– me translating that to others, that hasn't
14:43
Speaker 1: worked for them. But it works in my brain.
14:46
Speaker 2: I'll admit that. That element of prediction versus projection versus
14:55
Speaker 2: like all of that is a different way of looking
14:58
Speaker 2: at the world, which means you probably do this all
15:01
Speaker 2: the time. Weather is actually, you know, it's a guess
15:05
Speaker 2: in some cases. but yet we look at our weather
15:09
Speaker 2: app and we're like, is it going to rain today?
15:10
Speaker 2: And you're looking outside, you're like, well, it's clearly raining,
15:13
Speaker 2: but yet I'm still assuming that my weather app is correct.
15:17
Speaker 2: I just think it's the way we live our world.
15:19
Speaker 2: In fact, there's been some studies that have shown we
15:21
Speaker 2: are very bad as people. We are very bad at
15:25
Speaker 2: understanding what that means in terms of risk prediction, 99%
15:30
Speaker 2: versus 80%. Like, what does that actually mean if it
15:33
Speaker 2: comes to my day-to-day existence? We are really bad at that.
15:37
Speaker 1: So what is teaching people? I mean, the thing is,
15:41
Speaker 1: is that there's one thing to figure out how to
15:43
Speaker 1: teach this to school children. It's another thing to figure
15:47
Speaker 1: out what to do, what you have to do in
15:49
Speaker 1: higher ed and preparing for the next generation of jobs.
15:51
Speaker 1: But we've got an entire population that, whether they like
15:55
Speaker 1: it or not, needs to understand how this works. So
15:58
Speaker 1: how do you have this sort of group learning exercise
16:01
Speaker 1: that we're in the middle of right now?
16:03
Speaker 2: So it's about the practical knowledge. It's about the practicality
16:06
Speaker 2: of what is it that I can do personally today
16:09
Speaker 2: that allows me to understand my nuances, my overtrust. I
16:13
Speaker 2: talk a little bit about actually a lot about overtrust
16:16
Speaker 2: of AI and understand how do I retrain myself based
16:20
Speaker 2: on my day to day. One example, we used to
16:23
Speaker 2: kind of laugh when we talked about, well, I'm going
16:25
Speaker 2: to have a media off day. So I'm not going
16:29
Speaker 2: to look at any of my social media. I'm not
16:31
Speaker 2: going to answer my cell phone or text. And we
16:34
Speaker 2: pride ourselves on that. But we can do the same
16:36
Speaker 2: thing with respect to artificial intelligence and retrain and understand
16:41
Speaker 2: what it is. If I'm doing search, now I have
16:43
Speaker 2: an AI summarizer for all of that. OK, let's try
16:47
Speaker 2: to do a little bit different. take three different search engines,
16:50
Speaker 2: which we used to do, take three different search engines,
16:54
Speaker 2: see how the information is different. You will be surprised.
16:58
Speaker 2: So now you're starting to think about, oh wait, AI,
17:00
Speaker 2: there is no general AI. Like each one is different,
17:03
Speaker 2: just like people. They give you bad advice all the time,
17:07
Speaker 2: three different ways.
17:08
Speaker 1: I do this a lot. I use three models and
17:12
Speaker 1: I always never let one model have the final say.
17:17
Speaker 1: I you work the models against each other. Is this
17:23
Speaker 1: research complete? What's missing? Vice versa. And it's funny, what
17:29
Speaker 1: that tells me is that AI is not going to
17:31
Speaker 1: have the, you know, the fact that there are, I
17:34
Speaker 1: want to say slightly different outcomes. They're generally 85% overlap, right?
17:38
Speaker 1: But it is the 15% that makes it either really
17:43
Speaker 1: good or really sloppy, right? The difference between AI slop
17:47
Speaker 1: and really good AI tools is that 10% to 15%,
17:53
Speaker 1: which is if we already know that no one model
17:57
Speaker 1: has the answer, then are we overselling this a little bit? Well, so-
18:08
Speaker 1: I would say we, we being the collective society.
18:13
Speaker 2: We are, I would say overselling based on the fact
18:18
Speaker 2: that we are selling it as if it is perfect.
18:23
Speaker 2: But the fact is, is that it is better than
18:25
Speaker 2: what we had three years ago. Again, I'll use search, right?
18:30
Speaker 2: I remember trying to do a search and I'd have
18:33
Speaker 2: to have four different, I'm going to call them problems,
18:35
Speaker 2: but four different search criteria to get to the answer
18:39
Speaker 2: that I was actually looking for. Or I'd have to
18:41
Speaker 2: go through like 10 links and be like, you know what?
18:44
Speaker 2: This is ridiculous. Let me change my search, like my
18:47
Speaker 2: search terms now. I have shortened my time by using
18:52
Speaker 2: the summarizer, even though I still use different types of
18:55
Speaker 2: search engines. So what does that mean? One, I'm more
18:59
Speaker 2: efficient with my time. I get to the answer faster,
19:03
Speaker 2: but I've already trained myself to basically, one answer is
19:07
Speaker 2: not the truth. I still do a comparison function.
19:09
Speaker 1: Right.
19:10
Speaker 2: So it's better than going through all the links and
19:13
Speaker 2: clicking on them and navigating and then coming back and
19:15
Speaker 2: coming to trash like, oh, I hope my kid didn't
19:18
Speaker 2: see that page.
19:18
Speaker 1: Right.
19:19
Speaker 2: Like all of those things. Right. It's a lot easier.
19:25
Speaker 1: This episode of the Chuck Todd cast is brought to
19:27
Speaker 1: you by Kalshi. It's an exchange where you can trade
19:31
Speaker 1: on what you actually think is going to happen. Sports, politics, culture,
19:35
Speaker 1: whatever's on your mind. It's it's it's a market. You know,
19:39
Speaker 1: I think of it as a share of stock in
19:41
Speaker 1: a person or idea, if you will. So, for instance,
19:45
Speaker 1: you can trade in a market of blue wave versus
19:48
Speaker 1: blue tsunami in the midterms. on Kalshi. The odds, in fact,
19:52
Speaker 1: on Kalshi right now of a blue wave, and that
19:54
Speaker 1: is defined by Democrats holding 218 or more House seats
19:59
Speaker 1: and 49 or more Senate seats, and they have the
20:03
Speaker 1: blue wave at 81%, which means for essentially, if you
20:07
Speaker 1: don't believe, Democrats are going to sweep. or you don't
20:10
Speaker 1: believe they're going to get to 50 Senate seats and
20:13
Speaker 1: 218 House seats, it costs you only 19 cents per share. Now,
20:18
Speaker 1: how do they define blue tsunami? Well, the odds of
20:21
Speaker 1: a blue tsunami where Democrats hold 235 or more House seats,
20:24
Speaker 1: and remember, they haven't been over 235, I believe, this century,
20:29
Speaker 1: or 51 plus Senate seats. The blue tsunami scenario, a
20:33
Speaker 1: share of that costs you 43 cents, which means a
20:37
Speaker 1: majority of of this market believes it is not going
20:42
Speaker 1: to be a blue tsunami. The way an investor would
20:45
Speaker 1: look at this, if you're pretty convinced there's going to
20:47
Speaker 1: be a blue tsunami, but you want to hedge a
20:49
Speaker 1: little bit, right? What you're really predicting is that the
20:54
Speaker 1: poles are off, right? And if you believe the poles
20:56
Speaker 1: are saying this is going to be a blue wave,
20:59
Speaker 1: but if you believe the poles are off, maybe you
21:01
Speaker 1: believe they're more likely to be off on the, you know,
21:03
Speaker 1: that it's going to be a blue tsunami. But if
21:06
Speaker 1: the polls are off the other way, then it's a
21:08
Speaker 1: blue mirage, right? The fact is, this is a pretty
21:10
Speaker 1: good investment you could make here where you're hedging. You're
21:14
Speaker 1: not risking as much as you could if you decide
21:17
Speaker 1: to trade on the blue tsunami. And at the same time,
21:21
Speaker 1: you hedge that trade with a no on the blue wave.
21:26
Speaker 1: But I could tell you right now, blue tsunami seems
21:29
Speaker 1: like a discount. And so does no on the blue wave.
21:33
Speaker 1: So here's what you need to know on CalSheet. You're
21:35
Speaker 1: going up against other traders directly. There's no house that's
21:38
Speaker 1: setting the odds, right? It's just people putting their opinions
21:42
Speaker 1: on the line and that's what determines this market. Trade
21:46
Speaker 1: $ 25 with the code ChuckTodd and they'll throw in another
21:50
Speaker 1: $ 25 instantly. Please remember to trade responsibly. This gets to, though,
21:58
Speaker 1: the issue of what of, you know, a large language
22:01
Speaker 1: model is only good as the is what it trains on. Right.
22:05
Speaker 1: The data that it trains on. And like I said,
22:08
Speaker 1: I've had a lot of authors, particularly African-American authors on
22:12
Speaker 1: this program about what a forgotten history, missing history, lost history.
22:16
Speaker 2: Right.
22:17
Speaker 1: We know we know what it is, the whitewashed history.
22:20
Speaker 1: If we have if we're just going to summarize history.
22:24
Speaker 1: You know, I say this, it doesn't, there's a lot
22:27
Speaker 1: of groups of what I would say were sort of
22:30
Speaker 1: minority groups pushed aside over the last 500 years, whether
22:34
Speaker 1: it's Jewish people, African-Americans, Islamic people, indigenous folks, whose histories
22:43
Speaker 1: really only in the last 30 years are being resurfaced, right?
22:48
Speaker 1: Are getting a little bit of attention you know, that's,
22:55
Speaker 1: you know, do we embed in our stereotypes and, and
22:59
Speaker 1: how do we, you know, this for me feels like
23:02
Speaker 1: the real challenge of making these trusted and effective tools.
23:08
Speaker 2: Yeah. So, you know, I always ask the question of
23:11
Speaker 2: what is the purpose? What is the end goal of
23:14
Speaker 2: what we're doing? So large language models, um, if we
23:18
Speaker 2: are looking at, um, the history of communities. We know,
23:23
Speaker 2: for example, Central Park used to be the Seneca Village.
23:26
Speaker 2: It used to be a black neighborhood, middle class, that
23:29
Speaker 2: got wiped out to create Central Park in New York.
23:33
Speaker 2: A lot of people don't know that. And so how
23:36
Speaker 2: do we put that in if we're thinking about issues
23:39
Speaker 2: with redlining or changing the voting map? That's the history
23:44
Speaker 2: that can allow us to see, oh, wait, there's something
23:47
Speaker 2: historically wrong of what we have now. How do we
23:50
Speaker 2: get to the next stage? But if it's around math,
23:53
Speaker 2: two plus two equals four. I'm not quite sure. There
23:57
Speaker 2: might be history that's lost in terms of math and
24:00
Speaker 2: how we look at it, but we all accept that
24:02
Speaker 2: this is the math that we're using today, even though
24:05
Speaker 2: we know, you know, the abacus might be different in
24:09
Speaker 2: terms of math.
24:10
Speaker 1: You're speaking like an engineer. You're basically saying, hey, look,
24:13
Speaker 1: for essentially everything that's number based to loosely, this is
24:18
Speaker 1: going to be an incredibly exponentially incredible tool. Where we're
24:24
Speaker 1: going to consistently get into fights is what we've been
24:26
Speaker 1: fighting about since the dawn of us writing things down, right?
24:31
Speaker 1: Which is the subjectivity of human interpretation.
24:35
Speaker 2: Yeah, and this is one of the things that I
24:37
Speaker 2: keep talking about, which is the value of humanities, the
24:42
Speaker 2: social studies, linguistics, creative, the arts, right? Because that is
24:48
Speaker 2: the practice that is allows us to figure out how
24:51
Speaker 2: to become critical around things that we read or things
24:55
Speaker 2: that we hear. We look at it through that lens
24:58
Speaker 2: of I have two authors that write about the same event.
25:01
Speaker 2: And yet, because they have different lived experiences, their narration
25:05
Speaker 2: around it is different. We're good in the humanities and
25:10
Speaker 2: the social sciences. They're really good at getting students to
25:14
Speaker 2: learn and question that, which is why I think they
25:17
Speaker 2: have a lot of value in this age of AI
25:20
Speaker 2: because that's their discipline and they can teach us a
25:23
Speaker 2: lot about how to be critical if we're reading an
25:26
Speaker 2: AI summarization and saying, hey, how do I critique this
25:31
Speaker 2: commentary right here? What is the discipline to do that
25:35
Speaker 2: and the practice to get us into that mode?
25:39
Speaker 1: So let's talk about the challenge of teaching math now
25:42
Speaker 1: going forward, right? One of the things I remember seeing
25:46
Speaker 1: this in a focus group of some parents a few
25:49
Speaker 1: years back, and this was parents complaining a bit more
25:52
Speaker 1: about social media and tech. This was really just before
25:55
Speaker 1: the AI, now that AI has sort of become synonymous
25:59
Speaker 1: with our concerns about tech. But it was, you know,
26:03
Speaker 1: I don't know what the skill sets my kids are
26:07
Speaker 1: going to need. You know, it's like, You know, I
26:11
Speaker 1: think my parents' generation, I'm in my 50s, I feel
26:14
Speaker 1: like my parents' generation thought, well, they kind of had
26:16
Speaker 1: an idea of what skills I needed to have to
26:18
Speaker 1: survive in my adulthood. Towards the end of my kids' schooling,
26:27
Speaker 1: one is out of college, one's in the middle of college,
26:29
Speaker 1: I found myself, ooh, I don't know what skill sets
26:33
Speaker 1: they're learning now that'll be obsolete when they're 40, right?
26:38
Speaker 1: So math is a good one. If we're solving every
26:42
Speaker 1: math problem now, right, which apparently I think we're taping
26:46
Speaker 1: middle of October, I believe there was another, it feels
26:49
Speaker 1: like every week there's another barrier broken in the world
26:52
Speaker 1: of mathematics. Hey, we finally solved this problem. AI is
26:56
Speaker 1: going to, you know, we're going to solve every problem.
27:00
Speaker 1: So how do you teach math and what should be,
27:02
Speaker 1: and I think this is something I'm guessing we're all,
27:06
Speaker 1: particularly those of you in the engineering space, like What
27:09
Speaker 1: is the skill set you need in order to get
27:12
Speaker 1: the most out of AI?
27:14
Speaker 2: So I think one fundamental skill set is the ability
27:18
Speaker 2: to have critical thinking. Because what we teach in terms
27:23
Speaker 2: of the tools today will be obsolete 20, 30, 40 years.
27:27
Speaker 2: I remember growing up when I was in college, the
27:30
Speaker 2: professors would ask about the, why don't we use the
27:32
Speaker 2: slide ruler anymore?
27:33
Speaker 1: Did you have one? My grandfather was a mechanical engineer.
27:37
Speaker 1: I was going to ask you about the slide rule.
27:39
Speaker 1: And I remember when I turned 12 or 13, he goes,
27:43
Speaker 1: let me, he gave me a nice slide rule. He
27:46
Speaker 1: spent a lot of money on this. It's like in
27:47
Speaker 1: the early 80s. just when computers were just, but we
27:51
Speaker 1: had those fancy calculators. And by then you were using
27:56
Speaker 1: to do trigonometry or any of that, you know, the
27:59
Speaker 1: cosine stuff. I'm really stretching back to my days there, right?
28:03
Speaker 1: A little bit on trig 101. But, and I remember
28:07
Speaker 1: going to my dad, I said, you know, I don't
28:10
Speaker 1: know if we're ever going to use a slide rule.
28:12
Speaker 1: They told us we had to buy a calculator. One
28:14
Speaker 1: of those fancy scientific, I think back then we used
28:16
Speaker 1: to call them scientific calculators.
28:18
Speaker 2: Yes.
28:20
Speaker 1: And it made the slide rule obsolete. Yet my grandfather thought, well,
28:23
Speaker 1: how do you not understand how a slide rule works?
28:25
Speaker 1: You're not going to understand engineering if you don't understand
28:28
Speaker 1: a slide rule.
28:29
Speaker 2: Exactly. And I was in that era where the professors
28:33
Speaker 2: were lamenting because we have moved to the calculators, the
28:36
Speaker 2: scientific calculators. And now in engineering, they're using computers, right?
28:41
Speaker 2: So they're using computes. They're still building bridges. They're still
28:45
Speaker 2: building rocket ships. They're still programming computers. And so it's shifted.
28:50
Speaker 2: But one of the things that we as educators do
28:53
Speaker 2: is how do we teach you how to think? I
28:57
Speaker 2: think that's really what it is. Like even with the calculator,
29:00
Speaker 2: scientific calculator, there was still some things you had to do,
29:02
Speaker 2: fundamental principles, before you could even figure out what to
29:05
Speaker 2: put in the calculator to get the answer. It's really
29:08
Speaker 2: that element because I think of it for computing. I
29:14
Speaker 2: know probably about 20 different languages in computing. We don't
29:20
Speaker 2: really do Fortran anymore. We don't really do Pascal anymore. COBOLs,
29:27
Speaker 2: every so often. I mean, these are people, like my
29:28
Speaker 2: students have no idea. What is Fortran?
29:30
Speaker 1: Are you even going to bother teaching different coding languages?
29:34
Speaker 2: That's the whole thing, no. But if you teach one
29:38
Speaker 2: and get them to think through the process of how
29:40
Speaker 2: do you learn, it means that in five years or
29:43
Speaker 2: 10 years, when is the next, you know, whatever it is,
29:46
Speaker 2: list 5.0 coupled with generative AI. I don't know. They'll
29:51
Speaker 2: be like, oh, I can learn how to use this
29:53
Speaker 2: tool to figure out how to program with my AI
29:57
Speaker 2: in conjunction with my AI.
29:59
Speaker 1: Well, I think about the last 20 years. I mean,
30:01
Speaker 1: I would say when I was raising my kids in
30:04
Speaker 1: the late aughts, early teens, everybody said, well, they've got
30:08
Speaker 1: to learn to code. And it was this idea in
30:12
Speaker 1: the same way that my mom in Miami said, you
30:15
Speaker 1: need to learn Spanish, right? It was the same mindset. Like, hey,
30:19
Speaker 1: this is the world you're going to grow up in.
30:21
Speaker 1: You need to speak the language of the world you're
30:23
Speaker 1: growing up in, whether it was the physical world of
30:27
Speaker 1: South Florida, Miami, where I grew up with Spanish versus
30:30
Speaker 1: the technical world of this. And then it turns out,
30:34
Speaker 1: Guess what? Between Google Translate, you know, I think we're
30:37
Speaker 1: about five years away from never having to learn a
30:40
Speaker 1: foreign language because it's all going to be instantly translated
30:44
Speaker 1: to code. So what have we done over the last
30:48
Speaker 1: 20 years? Did we waste 20 years of education on kids?
30:52
Speaker 2: I think we didn't waste education on all kids, but
30:58
Speaker 2: we have to frame how we think about education now.
31:02
Speaker 2: Coding as an example, it was a big thing. Coding, coding, coding, coding.
31:06
Speaker 2: And now if you look at the top jobs that
31:08
Speaker 2: are being eradicated by AI. We did magnet schools for.
31:10
Speaker 1: Coding, remember? Like we were doing charter schools for coding, right?
31:14
Speaker 2: I was there too. I did camps and summer camps
31:17
Speaker 2: for students to learn coding. But what I would say
31:21
Speaker 2: is how you think about what is it that I
31:25
Speaker 2: need to do today in terms of the basic skills?
31:28
Speaker 2: Because the fact is you still have to get a job.
31:30
Speaker 2: So what are the jobs of today? Because you still
31:33
Speaker 2: have to have the ability to have a paycheck, feed
31:36
Speaker 2: your family, feed yourself, buy housing. So you have to
31:39
Speaker 2: have the skills for today. But as the jobs navigate
31:43
Speaker 2: and the jobs grow and we get to the next stage,
31:47
Speaker 2: can you learn those new jobs? Prompt engineering. Do you
31:51
Speaker 2: remember when prompt engineering came out as a job right
31:54
Speaker 2: after basically the chat GPT came out?
31:59
Speaker 1: Wow, that's what we're calling that? Prompt engineering?
32:02
Speaker 2: It was called prompt engineering. And you can go on,
32:05
Speaker 2: and they were hiring prompt engineers for over 100K.
32:08
Speaker 1: Just knowing how to use better words to get more
32:11
Speaker 1: out of the AI? Is that basically what you're describing?
32:14
Speaker 2: Yes. Yes. And now that's not a thing, but because
32:18
Speaker 2: it was so new. And I remember at that stage,
32:21
Speaker 2: the individuals who were being hired were the ones who
32:24
Speaker 2: understood language, understood conversation, understood all these little things to
32:29
Speaker 2: get the AI to get the answer correct. And they
32:31
Speaker 2: use the data to retrain the models. So now the
32:35
Speaker 2: AI was better at responding to us.
32:40
Speaker 1: So I was at a talk. My son attends SMU.
32:44
Speaker 1: And their university president is a guy named Jay Hartzell.
32:47
Speaker 1: He used to be at UT Austin. And he recently,
32:50
Speaker 1: I view him as somebody who was like, I can't
32:54
Speaker 1: do the public institution. Let me go private where there's
32:57
Speaker 1: fewer people minding my business type of mindset. But he
33:01
Speaker 1: said something interesting about what he thought the impact of
33:04
Speaker 1: AI would be on higher ed. And he said, and
33:07
Speaker 1: he's done this so far, is he wants to start
33:09
Speaker 1: to steer SMU away from STEM. not in a negative way,
33:14
Speaker 1: but that essentially everything goes back to basics that the
33:18
Speaker 1: quote unquote classical education, right? What was considered a curriculum
33:23
Speaker 1: of the late 19th century, early 20th century is essentially
33:28
Speaker 1: the bill that we're back to the idea that we're
33:31
Speaker 1: doing building blocks so that it's almost tech proof. Um,
33:36
Speaker 1: is that realistic?
33:38
Speaker 2: Um, I do not think so. There is some elements
33:41
Speaker 2: of classical education, but you have to realize classical education,
33:45
Speaker 2: it was really about being a finishing school. Like, let's
33:48
Speaker 2: really be real around how universities started. Universities were starting,
33:52
Speaker 2: it was pretty much the elite, and I need to
33:55
Speaker 2: send my sons and daughters to school so that they
33:58
Speaker 2: can take control. To be refined. It was a finishing school.
34:01
Speaker 2: And this was the intellectual discourse and all of these things.
34:05
Speaker 2: So there's some element of that But we do not
34:09
Speaker 2: have that hierarchy quite, I mean, frankly, it's changed. During
34:14
Speaker 2: the World Wars, the universities then became much more around, oh,
34:19
Speaker 2: we need a workforce. We need individuals that actually work,
34:22
Speaker 2: not the aristocrats, not the finishing schools. We need people
34:25
Speaker 2: that can actually think and go out in the field
34:28
Speaker 2: and be engineers and seamstresses and dot, dot, dot. So
34:32
Speaker 2: it's shifted and GI Bill, we're going to pay for
34:34
Speaker 2: people to go to school because they're not in the
34:37
Speaker 2: elite class. I think we are trying to redefine and
34:41
Speaker 2: figure out what we are now. We have a lot
34:44
Speaker 2: of universities. A lot of universities will close because we
34:47
Speaker 2: frankly have too many, unfortunately. It'll be the ones that last.
34:53
Speaker 2: Spelman College, for example, we will continue. But some of
34:56
Speaker 2: the smaller ones will struggle until they close because we
34:59
Speaker 2: haven't figured out what is our identity at this point.
35:03
Speaker 1: That's interesting you say that. Look, I think there's the
35:05
Speaker 1: demographic challenge, number one, that seems to be the big.
35:09
Speaker 1: But you think we have too many universities. How did
35:11
Speaker 1: we end up with too many universities, you think?
35:15
Speaker 2: One, we did have a large population. population of individuals
35:19
Speaker 2: going to college. It became this thing of you have
35:21
Speaker 2: to go to college to get a job. And trade
35:24
Speaker 2: schools became less of an interest in a lot of cases.
35:29
Speaker 2: And I think trade schools have a great value in,
35:31
Speaker 2: especially in the U.S. and other places. But we sort
35:34
Speaker 2: of moved away from trades and went to, you have
35:37
Speaker 2: to go to college. And the problem is, is that
35:41
Speaker 2: it was a college without necessarily an output. You go
35:44
Speaker 2: to college, get a degree in And X didn't necessarily
35:50
Speaker 2: associate with a job. It was just, you go to college.
35:54
Speaker 2: So a lot of universities and colleges popped up just
35:57
Speaker 2: to address the X. You had this influx of students
36:00
Speaker 2: that just wanted a college degree. And now we are here.
36:04
Speaker 2: We have this demographic cliff where we have less 18-year-olds
36:09
Speaker 2: going to college. And you have colleges that haven't been
36:12
Speaker 2: able to describe what that X is. What is the
36:15
Speaker 2: value of X? And it costs. So what is the
36:18
Speaker 2: value of X associated with the cost it takes to
36:22
Speaker 2: achieve and receive whatever the X degree is?
36:27
Speaker 1: So give me that answer for Spellman. How are you
36:31
Speaker 1: addressing that question?
36:32
Speaker 2: Yeah, so Spelman is, I will say, is a boutique university.
36:37
Speaker 2: We take the creme de la creme. We focus on
36:40
Speaker 2: Black women excellence. But we really think about what is
36:44
Speaker 2: that hybrid of the, would you say, classical education, the
36:50
Speaker 2: social scientists, humanities, the creative, but couple it with what
36:55
Speaker 2: is it that you need to have skills moving forward.
36:58
Speaker 2: We're the number one producer, for example, of Black women
37:00
Speaker 2: who go into STEM PhDs. the number of lawyers and
37:05
Speaker 2: judges that we have that are out, the number of
37:07
Speaker 2: C-suite executives that are in our, you know, 40, 50,
37:12
Speaker 2: 60 year olds. We have an amazing ability to train
37:16
Speaker 2: these leaders, these women leaders, Black women leaders for that world.
37:20
Speaker 2: But we are not, we have, you know, a class
37:23
Speaker 2: of 650-ish that come in. We are not a flagship
37:28
Speaker 2: where we bring in tens of thousands of students. But
37:32
Speaker 2: that is our focus and that's our emphasis. And so
37:35
Speaker 2: the way we think about it is we are not
37:36
Speaker 2: going to be everything to everyone. We have an identity.
37:40
Speaker 2: We have a legacy. Those are the schools, I believe,
37:43
Speaker 2: that are going to last. It's going to be the flagships.
37:45
Speaker 2: They have a unique identity.
37:46
Speaker 1: Now, that makes sense. I guess I'd love to. So
37:52
Speaker 1: what you just described and what you really described of
37:54
Speaker 1: what happened to universities post GI Bill is we took
37:58
Speaker 1: we took the trade school and we took the finishing
38:01
Speaker 1: school and we merged them, right? Essentially. In that you
38:05
Speaker 1: go to college both to sort of learn how to be,
38:09
Speaker 1: you know, a better human, right? A little more empathetic,
38:12
Speaker 1: a little how to live with others, how to work
38:14
Speaker 1: with others. But you also then had to pick a major,
38:17
Speaker 1: pick a trade, essentially, right? What are you going to do?
38:20
Speaker 1: What are you going to do for a living? And
38:23
Speaker 1: then over the last, you know, and that led over
38:25
Speaker 1: the last 20 years of, you know, take my there's
38:29
Speaker 1: a small little catholic school in the neighborhood i live
38:31
Speaker 1: in hey they're no longer going to offer english as
38:34
Speaker 1: a major anymore right because the idea was what is
38:37
Speaker 1: english what is what do you what job is directly
38:42
Speaker 1: associated with that besides english teacher or writer like what
38:47
Speaker 1: is it and we we went to the trade do
38:50
Speaker 1: you think we should unwind that a little bit um
38:54
Speaker 1: or actually focus more on the trade side of things
38:57
Speaker 1: when it comes to higher education?
38:59
Speaker 2: So I think we need to do both and. I
39:04
Speaker 2: think that when students, I always call it the headwinds,
39:08
Speaker 2: students are still attuned. They've learned, parents have learned that
39:14
Speaker 2: you go to college with the ability to bring wealth
39:19
Speaker 2: back to your community. It's a source of finding your
39:23
Speaker 2: passion and getting paid for it. We can't change that messaging.
39:28
Speaker 2: That messaging is there. We can't suddenly say, oh, by
39:30
Speaker 2: the way, the messaging for the last 50 years is wrong.
39:34
Speaker 1: No, we've learned that if you're interested in what you're doing,
39:37
Speaker 1: you're going to be better at your job and there's
39:38
Speaker 1: probably going to be more advancement for you and maybe
39:42
Speaker 1: the world you may advance the cause of what you're
39:46
Speaker 1: working in because you enjoy it.
39:47
Speaker 2: Yes, because you enjoy it. And so I think we
39:50
Speaker 2: need to, that messaging needs to be much clearer. Saying
39:55
Speaker 2: it's one, aligning your passion to the job, but also
40:01
Speaker 2: showing that the classical education, that the reason why you're
40:05
Speaker 2: learning literature, the reason why you're learning history, the reason
40:09
Speaker 2: why you're learning even some fundamental chemistry and math is
40:13
Speaker 2: so that you are enabled to figure out your job
40:17
Speaker 2: later on, 20, 30, 40 years. And I will tell you,
40:21
Speaker 2: you interact with most alumni. At Spelman, I interact with
40:27
Speaker 2: quite a bit. The job they have when they're 50
40:30
Speaker 2: was not the job or the type of job they
40:32
Speaker 2: had when they graduated at 21, 22. And they have
40:36
Speaker 2: the confidence to change and switch. Like, how do we
40:39
Speaker 2: do that? But then also there are trade schools. So
40:42
Speaker 2: if a student says, you know what, I really want
40:44
Speaker 2: to be a railroad engineer. That's really what I want
40:48
Speaker 2: to do. Well, there's not a lot of railroad engineering
40:52
Speaker 2: programs in four-year colleges. And so let's not just say, well,
40:58
Speaker 2: you can be any engineer. Just go to college and
41:00
Speaker 2: you'll be an engineer and you can figure it out.
41:02
Speaker 2: It's like, okay, let's have a real conversation around, do
41:06
Speaker 2: you want to be on the railroads? Is this a
41:08
Speaker 2: tech skill, a tech trade? Or do you want to
41:10
Speaker 2: be an engineer that happens to work for a railroad company?
41:16
Speaker 1: What do you want to see in K through 12 reform?
41:20
Speaker 1: Or frankly, I call it pre-K through 12 reform. And
41:23
Speaker 1: I personally think we need to start looking at it
41:25
Speaker 1: in smaller chunks. Like pre-K through third grade is one
41:30
Speaker 1: thing that we need to focus on. And then I
41:32
Speaker 1: think once you get there, it becomes easier to figure out, okay,
41:36
Speaker 1: what is the path going forward? How much is it
41:39
Speaker 1: a trade path? How much is it an intellectual path?
41:41
Speaker 1: How much is it a skills training path in different things,
41:46
Speaker 1: but I feel like we need to do a dramatic
41:50
Speaker 1: sort of rethinking of our public education system. I hate
41:53
Speaker 1: what's happened with vouchers in charter schools. I'm not going
41:56
Speaker 1: to bring you into a political argument on that, but
41:58
Speaker 1: I think the voters are speaking pretty loudly about how
42:01
Speaker 1: they feel about vouchers because wherever it's been put in,
42:05
Speaker 1: it's not popular. What makes a better student at Spelman?
42:11
Speaker 1: What's missing right now in public education that would give
42:15
Speaker 1: you the ability to do something a little bit different
42:19
Speaker 1: at Spelman if you chose to?
42:21
Speaker 2: Yeah, so I think what's missing, honestly, is this focus
42:26
Speaker 2: on impacting your community, impacting society. I will tell you,
42:31
Speaker 2: when our students start here at Spelman, I talk to them.
42:35
Speaker 2: They volunteered at the local Red Cross or the hospitals
42:39
Speaker 2: or they built up a fresh garden in their neighborhood.
42:42
Speaker 2: Like they all have a story about something they did
42:46
Speaker 2: outside of themselves that shows that they care about others.
42:53
Speaker 1: This episode of the Chuck Todd cast is brought to
42:55
Speaker 1: you by Quince. You know, the best pieces in your
42:57
Speaker 1: closet are the ones you can count on, whether you're
43:00
Speaker 1: heading to work, going out or keeping things a bit casual.
43:03
Speaker 1: And this time of year, Quince makes finding something great
43:06
Speaker 1: for everyone else just as easy. The holiday season is coming.
43:10
Speaker 1: That's right. It's we're under 90 days, folks. It's where
43:14
Speaker 1: I'm shopping for myself and everyone else on my list
43:17
Speaker 1: because I know Quince delivers. Nobody's ever disappointed in stuff
43:22
Speaker 1: they get from Quince.
43:23
Speaker 2: Why?
43:23
Speaker 1: Quince focuses on high quality staples made with premium materials
43:27
Speaker 1: like 100% Mongolian cashmere, 14 karat gold and Italian leather.
43:32
Speaker 1: Their cashmere sweaters start at just 60 bucks. And the
43:36
Speaker 1: lineup doesn't stop there. They have premium down jackets, wool outerwear,
43:40
Speaker 1: cashmere accessories, everything you and your loved ones need to
43:43
Speaker 1: stay warm and look put together all season long without
43:47
Speaker 1: breaking the bank. Everything at Quince is priced 50 to 80%
43:51
Speaker 1: less than similar brands. How do they do this? Well,
43:53
Speaker 1: they work directly with factories and not just any factories,
43:58
Speaker 1: ethical factories. They cut out the middleman or the middle person,
44:02
Speaker 1: And so you can kind of get in your sort
44:05
Speaker 1: of luxury quality pieces without paying that luxury price tag.
44:11
Speaker 1: You know, it sounds like ad copy, but Quince is
44:14
Speaker 1: responsible for sort of the stuff I just grab and go,
44:17
Speaker 1: the stuff that I know I can just throw on
44:20
Speaker 1: that works well. I got these two pairs of shorts
44:23
Speaker 1: from Quince. And, you know, the shorts are both warm.
44:26
Speaker 1: They're kind of khaki, but they're kind of not. They're
44:29
Speaker 1: just easy to wear. I'm wearing a pair of those
44:32
Speaker 1: shorts right now as I'm taping this podcast. So when
44:35
Speaker 1: it comes to gifting, Quince is a lot more than clothing, too.
44:38
Speaker 1: By the way, they've got fine jewelry, leather bags, hotel-quality bedding,
44:44
Speaker 1: chef-grade cookware. People love a nice piece of equipment to
44:49
Speaker 1: play with in the kitchen these days. Five-star luggage. Plenty
44:52
Speaker 1: of ways to find something special for everyone on your list.
44:56
Speaker 1: So give the gift. of quality this holiday season. Download
45:01
Speaker 1: the Quince app for app-exclusive offers or go to quince.com.
45:05
Speaker 1: You get free shipping on your order, 365-day returns. Now
45:09
Speaker 1: available in Canada and the UK, too. And when Quince
45:12
Speaker 1: asks where you heard about them, please let them know
45:15
Speaker 1: it was the Chuck Todd cast. It's the best way
45:19
Speaker 1: to support the show and the best way to make
45:21
Speaker 1: sure you're getting every discount. That's the student you're looking
45:27
Speaker 1: for though, isn't it? Don't you sort of like, if
45:29
Speaker 1: they're already doing that, you want them at Spelman, correct?
45:31
Speaker 2: We want them. And so why in public school, and again,
45:34
Speaker 2: we can talk about funding and it's still hard. Teachers
45:38
Speaker 2: are stretched, especially in certain communities in terms of teaching.
45:42
Speaker 2: And then you want me to be the faculty volunteer
45:45
Speaker 2: for these student projects. It becomes very difficult. But I
45:49
Speaker 2: truly believe that when you root your education around a
45:53
Speaker 2: cause and It actually shows students the reason for whatever
45:59
Speaker 2: you're learning. As an example, we have a program here called,
46:04
Speaker 2: believe it or not, cosmetic science, which is in chemistry. Like, oh, wait,
46:11
Speaker 2: hold it. Yes. Like a lot of women wear makeup
46:15
Speaker 2: as an example. You know, there's a science to that.
46:18
Speaker 2: There's a science to how it spreads. There's a science
46:20
Speaker 2: to if you have an allergic reaction. There's a science
46:23
Speaker 2: to how do you take two colors and you blend
46:25
Speaker 2: them together based on the molecules. It's actually chemistry at
46:29
Speaker 2: its root cause. Right. And so can we do that
46:33
Speaker 2: more in public education where you have a student, they're
46:36
Speaker 2: interested in music? All right, guess what? Music is actually math.
46:40
Speaker 2: Like if you think about the beats and all of this,
46:43
Speaker 2: can we infuse then math in music? So you're learning
46:46
Speaker 2: your music and you're learning your beats and you're learning
46:48
Speaker 2: how to put together an X, but we're teaching you
46:52
Speaker 2: the math behind it. So you're like, oh, I have
46:53
Speaker 2: to learn Wait, I have to learn fractions? Yeah. Oh,
46:57
Speaker 2: but then I can become a better beatmaster? All right.
47:00
Speaker 2: I'm there.
47:01
Speaker 1: Now I understand what... No, I mean, this is why
47:03
Speaker 1: it seems to me we've got to do this in
47:05
Speaker 1: bite-sized chunks that, you know, it isn't a high school
47:09
Speaker 1: problem we have. I actually think it's a pre-K problem.
47:12
Speaker 1: And then once you get to universal pre-K through third, well,
47:15
Speaker 1: then you're probably going to raise the baseline. And then
47:19
Speaker 1: you have a better chance of designing a more individual...
47:23
Speaker 1: personalized public education experience going forward but if you're spending
47:28
Speaker 1: all your time playing catch-up all the time you don't
47:31
Speaker 1: get to do.
47:31
Speaker 2: That and that's what we're doing now we're playing catch-up
47:35
Speaker 2: and so and i would just say when i.
47:38
Speaker 1: Hear about this at a lot of universities that a
47:40
Speaker 1: lot of universities the freshman year is about bringing kids
47:44
Speaker 1: up to speed to get them prepared for college really
47:47
Speaker 1: your sophomore year is really you know your first year
47:49
Speaker 1: of college do you feel that way these days um I.
47:53
Speaker 2: Would say a little bit, but it's more around the
47:56
Speaker 2: learning skills, the study skills, the, you know, this is
48:02
Speaker 2: what an A looks like.
48:04
Speaker 1: Is it maturity, you think, or is it poorly or
48:07
Speaker 1: just poor schooling?
48:10
Speaker 2: I will tell you there's a difference between the students
48:13
Speaker 2: that learn through COVID and the students that didn't.
48:17
Speaker 1: Tell me about it.
48:19
Speaker 2: Well, so, you know, if I... Pull it back, the
48:23
Speaker 2: students who had not learned how to learn, the students who,
48:28
Speaker 2: the study skills and working in a group and a
48:31
Speaker 2: group project, if they didn't have those skills, which we
48:35
Speaker 2: do in college all the time, this is their first
48:39
Speaker 2: exposure to it because a lot of schools do not
48:41
Speaker 2: go back to that tradition because in middle school, They
48:46
Speaker 2: went into lockdown. So socialization in middle school is so important.
48:51
Speaker 2: So they didn't learn that. They were in high school.
48:53
Speaker 2: And so now you have high school teachers coming back
48:56
Speaker 2: in person and their students do not have the skills of,
49:01
Speaker 2: this is what group projects looks like. Okay, so then
49:03
Speaker 2: you have to, you teach to the middle. That's usually
49:06
Speaker 2: what teachers do. And now your middle has shifted and
49:09
Speaker 2: is different. And so what we've seen is then they
49:12
Speaker 2: come to college and the way that they interact with
49:15
Speaker 2: the world, the way that they learn is different because
49:19
Speaker 2: the students had to learn differently in high school and
49:24
Speaker 2: middle school. And I do not know what's going to
49:27
Speaker 2: happen when the students that were at the elementary school
49:30
Speaker 2: level actually get to college.
49:33
Speaker 1: Interesting. Yeah. What that looks like. It is interesting to
49:37
Speaker 1: me that we're finally starting to see improvements in third
49:40
Speaker 1: and fourth grade reading levels. And, well, that makes sense
49:44
Speaker 1: because guess what? They started school after COVID.
49:48
Speaker 2: Yes.
49:48
Speaker 1: If you're in third or fourth grade, right?
49:50
Speaker 2: But look at how many years that is before they
49:52
Speaker 2: get to college. So we have this whole generation of
49:55
Speaker 2: students that we as educators.
49:58
Speaker 1: So another decade of this that you're going to be
50:00
Speaker 1: dealing with it, right?
50:01
Speaker 2: Correct. Correct. Correct. And then education is gonna shift again.
50:04
Speaker 2: That's what I'm saying. There's a cycle. I mentioned NASA,
50:07
Speaker 2: like every 10 years it changes. It's the same as
50:11
Speaker 2: with education.
50:11
Speaker 1: All right, I wanna hit two more topics. One goes
50:15
Speaker 1: back to your book because I just feel as if
50:19
Speaker 1: none of these AI companies have enough people thinking about
50:22
Speaker 1: equity in the room. And I don't care whether it's
50:25
Speaker 1: gender equity, racial equity, religious equity, you name it. It is,
50:31
Speaker 1: they're plowing forward, you know, making advancements. And I go
50:37
Speaker 1: back to the movie Contact, which is Matthew McConaughey, Jodie Foster,
50:41
Speaker 1: where they spend half of the movie debating whether an
50:46
Speaker 1: atheist meeting the extraterrestrial is a good idea, which I
50:50
Speaker 1: thought was like, you know what? I thought that was
50:53
Speaker 1: a very realistic response. I know our politics. I know
50:56
Speaker 1: the way religion, you know, I'm probably more on the
50:59
Speaker 1: science side of things, but I understand the importance of it.
51:03
Speaker 1: And yeah, that's just our civilization at this moment. That's
51:09
Speaker 1: a big part of our civilization. I don't feel like
51:13
Speaker 1: AI has grabbed all of this. How do we force
51:16
Speaker 1: the equity upon the development of this?
51:19
Speaker 2: Yeah, so I think, and I do mention in the book,
51:22
Speaker 2: which is a blueprint for both the developers, but also
51:26
Speaker 2: everyone else that's not a developer of the power that
51:29
Speaker 2: you have thinking about this. So it's not just developers, right?
51:34
Speaker 2: It's we have a say in how it's developed and
51:36
Speaker 2: we have a say in how it's used. And we
51:39
Speaker 2: sort of, I would say as humans, we need to
51:42
Speaker 2: not be so trusting of these technologies and using it,
51:47
Speaker 2: even though in some cases it makes our life convenient. Right.
51:51
Speaker 2: And that's the problem. I think that the convenience allows
51:54
Speaker 2: us to be forgiving of its use because we're not
51:59
Speaker 2: willing to feel uncomfortable not using it. It just is.
52:05
Speaker 2: I always think about maps. You and I probably grew
52:08
Speaker 2: up where you had the triptych when you wanted to
52:11
Speaker 2: do cross country.
52:12
Speaker 1: Yeah.
52:13
Speaker 2: Oh, no.
52:13
Speaker 1: And you got this little, yeah, that little spiral notebook.
52:17
Speaker 2: You have a spiral notebook and you figure out how
52:19
Speaker 2: to maps. And now I will tell you, most people
52:22
Speaker 2: would not know how to get across their own city,
52:25
Speaker 2: even if they've driven it every day, because we're using
52:29
Speaker 2: these maps. whatever Waze or Google Maps or whatever is
52:33
Speaker 2: out there. Like one example would be, you know, take
52:37
Speaker 2: off a map day and like actually try to figure
52:41
Speaker 2: out how to navigate from one place to another on
52:44
Speaker 2: your own, looking at landmarks. And so there's these practices
52:47
Speaker 2: that we can invoke. The other is, is we can
52:51
Speaker 2: also provide feedback when a large language model, whatever your
52:56
Speaker 2: choice is, gives you a wrong answer. There's ways to
52:59
Speaker 2: provide your correction. And if we have a movement of
53:02
Speaker 2: all of us pushing back, guess what? The developers and
53:05
Speaker 2: the companies are like, oh, there's a large number of
53:07
Speaker 2: people that said this answer is wrong.
53:11
Speaker 1: Do you think it's inevitable that we'll have specialty large
53:14
Speaker 1: language models, not necessarily like specialty in that you'll have
53:18
Speaker 1: a health large language model, but specialty in that This
53:22
Speaker 1: one has a Christian identity and this one has a
53:26
Speaker 1: Muslim identity and this one has a Jewish identity or
53:29
Speaker 1: this large language model has a indigenous identity, an African-American identity.
53:33
Speaker 1: What's your sense? Are we going to have a bunch
53:37
Speaker 1: of those models? I think of it almost like the
53:39
Speaker 1: way we had a black press and a Latino press
53:44
Speaker 1: in this country. Do we see that or am I
53:47
Speaker 1: thinking almost too American? Um, I want to say, I
53:51
Speaker 1: hope so.
53:53
Speaker 2: And why I say that is we always talk about
53:55
Speaker 2: bias being negative, but it's not always negative. I want
53:59
Speaker 2: a healthcare robot that's actually trained on female based data. Yeah. Right.
54:06
Speaker 2: I like, I just do.
54:08
Speaker 1: That's my preference. A lot of people have a preference
54:11
Speaker 1: for my wife and I go to a, a healthcare
54:14
Speaker 1: practice where there is a woman doctor and a male doctor,
54:18
Speaker 1: and one of us is more comfortable. I mean, that
54:20
Speaker 1: is just normal. Most women are more comfortable with a
54:22
Speaker 1: woman doctor, most men, vice versa. So your large language model,
54:27
Speaker 1: having a bias towards your gender health makes a lot
54:30
Speaker 1: of sense.
54:31
Speaker 2: And what allows us to do though, is to make
54:34
Speaker 2: a choice, right? So I can choose what I want,
54:38
Speaker 2: the identity of my counselor. I may want a mental
54:42
Speaker 2: health counselor. I want to make sure that it understands
54:45
Speaker 2: my neighborhood. Give me that choice versus just give me a,
54:49
Speaker 2: here's the general model, take it or leave it. in
54:52
Speaker 2: terms of I'm very US centric, but there's a whole
54:55
Speaker 2: world outside of the US. So if I'm in Africa,
54:58
Speaker 2: I don't necessarily want to interact with someone who's definitely
55:04
Speaker 2: US centric, because it's just different culturally, language wise, what's yes, no,
55:09
Speaker 2: what's approved, what's disapproved in terms of behaviors. Let's create
55:15
Speaker 2: models that have that.
55:17
Speaker 1: That are sort of, I mean, that's where I feel
55:22
Speaker 1: The worst thing is in the world of politics and
55:27
Speaker 1: exit polling, right? I remember one time somebody saying the
55:29
Speaker 1: worst thing than no exit polls is one exit poll,
55:33
Speaker 1: meaning that everybody's relying on the same thing and you
55:36
Speaker 1: actually either want none or you want five so that
55:39
Speaker 1: you have competing– theories. And I think we should feel
55:43
Speaker 1: that way about large language models. We don't want three.
55:46
Speaker 1: We don't want to be talking about chat versus Claude
55:48
Speaker 1: versus Gemini. We want 300, 3000, 3 million, right?
55:53
Speaker 2: We do. I think so. And it's because humans are diverse. Like,
56:00
Speaker 2: we are so different. You take, I identify as a
56:04
Speaker 2: Black woman. You put another Black woman, we are actually
56:07
Speaker 2: quite different. Depends on our social economics, our training. Are
56:11
Speaker 2: they an engineer?
56:12
Speaker 1: When were you born? All those things. Yeah.
56:15
Speaker 2: So don't label us the same and say, oh, here
56:18
Speaker 2: you are. And I think that's the beauty of human
56:21
Speaker 2: diversity is that we are so different across all of
56:25
Speaker 2: our parameters. We should expect nothing less of our AI agents,
56:30
Speaker 2: our robots that we interact with.
56:31
Speaker 1: All right. Before I ask my final question, it's a
56:34
Speaker 1: little bit of a rabbit hole. Do you watch For
56:40
Speaker 1: All Mankind, that alternative history series on Apple about the
56:45
Speaker 1: Russians beat the Americans to the moon?
56:47
Speaker 2: No, I have not.
56:49
Speaker 1: Okay. Well, I won't get into, I won't ask that,
56:52
Speaker 1: but it was about sort of, so the basic premise
56:55
Speaker 1: of the show is we end up making more advancements
56:57
Speaker 1: in space because we lose the initial race to the moon.
57:01
Speaker 1: And because we lose, we're Americans. We throw even more money,
57:05
Speaker 1: more everything. And eventually we're basing the moon. We're in
57:09
Speaker 1: Mars by the 90s. It's it's it basically the premise
57:13
Speaker 1: of the show is instead of having a nuclear arms
57:16
Speaker 1: race with the Soviets, we have a space race. The
57:19
Speaker 1: space race continues because we don't we don't win it
57:23
Speaker 1: right away. We lose.
57:24
Speaker 2: So we're like, OK, we got it.
57:26
Speaker 1: We got it.
57:26
Speaker 2: We're still in the race. And.
57:29
Speaker 1: I say this because, let me ask you, put your
57:31
Speaker 1: NASA hat back on. Why haven't we based the moon?
57:37
Speaker 1: And why aren't we in Mars yet? Because it feels
57:40
Speaker 1: like we were making all these advancements. And I know
57:42
Speaker 1: we had a, look, I would argue the economic downturn
57:46
Speaker 1: of the 70s pulled a lot of that funding back
57:49
Speaker 1: and we had challenges and suddenly politically investing in space
57:53
Speaker 1: wasn't good politics. It was actually bad. Why don't, no, no, no, no,
57:57
Speaker 1: don't take, we gotta have social security and healthcare before
58:00
Speaker 1: we worry about going to Mars. I feel like we've
58:05
Speaker 1: had the ability to do these things sooner. We just
58:08
Speaker 1: haven't had the will. Is that your experience at NASA?
58:12
Speaker 1: That we just haven't had the financial will to do the,
58:15
Speaker 1: if we wanted to get to Mars 30 years ago,
58:18
Speaker 1: we probably could have?
58:20
Speaker 2: We probably could have, but every decision that we make
58:25
Speaker 2: has a cost benefit analysis. And the fact is, is
58:29
Speaker 2: the reason we got to the moon was it was
58:31
Speaker 2: a challenge and it was an existential threat. Right. If
58:35
Speaker 2: we did not get to the moon first, then the
58:39
Speaker 2: other power would be stronger than us and might annihilate us. Right.
58:42
Speaker 1: So it sounds like the AI spin, by the way.
58:45
Speaker 2: But that's true. And so when we talked about all
58:50
Speaker 2: of these nations are putting funding into AI, that is
58:54
Speaker 2: the reason why. And. The difference, though, is that back
58:59
Speaker 2: then we did have guardrails. I mean, we weren't going
59:03
Speaker 2: to do certain things like, you know, let's put a
59:06
Speaker 2: nuclear reactor on the moon. We might have had them
59:09
Speaker 2: on our rover. but we weren't going to just stick
59:13
Speaker 2: a nuclear plant on the moon, right? So there was
59:15
Speaker 2: still some guardrails understanding of how did we think about
59:17
Speaker 2: this in a little bit of a safety mechanism, even
59:20
Speaker 2: though we were going as pretty fast, I would say.
59:24
Speaker 2: We just don't have that oomph. Because the threat is
59:30
Speaker 2: not just the one. It's an imaginary threat. It's a feeling.
59:35
Speaker 2: It's a, we want to be first, but it's not
59:38
Speaker 2: a solid, like this is absolutely a threat to humanity.
59:44
Speaker 2: Although there's this conversation of a threat to humanity, which
59:47
Speaker 2: basically means is AI versus humans, which caters toward a
59:53
Speaker 2: different way of developing versus, you know, the U.S., Against,
59:58
Speaker 2: I don't know, the water.
1:00:01
Speaker 1: Yeah. Well, let me ask this. Now that there's a
1:00:07
Speaker 1: private industry in space, somewhat of a private industry, and
1:00:13
Speaker 1: it feels more hybrid-ish-y, right?
1:00:15
Speaker 2: Quasi.
1:00:16
Speaker 1: Yeah, that you still need, you know, it's sort of like,
1:00:20
Speaker 1: why is... Again, I'm not putting this in your words.
1:00:24
Speaker 1: These are my words. Why is Elon Musk enamored with
1:00:27
Speaker 1: whoever's president? Because at the end of the day, SpaceX
1:00:29
Speaker 1: needs a good relationship with government if it actually wants
1:00:32
Speaker 1: to do what it wants to do, right? And why
1:00:36
Speaker 1: does Jeff Bezos need a relationship with the president? Same
1:00:38
Speaker 1: thing if he wants Blue Origin to have some aspects
1:00:41
Speaker 1: of it. What's going to happen in the next 50
1:00:45
Speaker 1: years when it comes to Mars, when it comes to,
1:00:48
Speaker 1: you know, Do you expect us to colonize Mars in
1:00:51
Speaker 1: the next 50 years? And given your experience at NASA,
1:00:54
Speaker 1: how the technology is there and we can do it?
1:00:57
Speaker 2: I would say if we discover natural resources....
1:01:03
Speaker 1: Let's be a reason.
1:01:04
Speaker 2: Yeah, it has to be a reason or suddenly the
1:01:07
Speaker 2: earth becomes uninhabitable and we need to find another home.
1:01:11
Speaker 2: But it's one of the things that I I'm a scientist.
1:01:14
Speaker 2: I love asking questions just because I'm curious. We we
1:01:19
Speaker 2: have no longer have that. curiosity for curious sake. We
1:01:25
Speaker 2: kind of have to have a reason. I mean, even
1:01:27
Speaker 2: the discovery, rediscovery, even though it was already inhabited by
1:01:30
Speaker 2: people of the U.S., the motivation was around resources.
1:01:36
Speaker 1: Right. Yeah, we were there looking, you know, people came
1:01:39
Speaker 1: from Europe and Asia looking for stuff. They weren't just going, oh,
1:01:42
Speaker 1: look at this. You know, this is nice.
1:01:44
Speaker 2: This is nice. I mean, Australia, you know, they needed
1:01:46
Speaker 2: the land because they needed to send their, you know,
1:01:50
Speaker 2: we need to send the people we don't want near
1:01:52
Speaker 2: us to Australia. It's how we're wired. So Mars has
1:01:58
Speaker 2: to have that kind of either we find aliens. So
1:02:02
Speaker 2: now it's like, oh, we need to get there because
1:02:04
Speaker 2: they may destroy us or there's natural resources. We need these.
1:02:08
Speaker 2: We no longer have whatever it is, water and minerals
1:02:11
Speaker 2: or things that we need to build stuff, our chips, maybe,
1:02:14
Speaker 2: for example. Yeah.
1:02:16
Speaker 1: So you expect both the moon and Mars. It depends
1:02:20
Speaker 1: on what resources we do find. depends on how much
1:02:23
Speaker 1: energy we put into colonizing them or basing them or
1:02:26
Speaker 1: whatever it is we want to call it.
1:02:28
Speaker 2: It has to have a return on investment. And the
1:02:30
Speaker 2: question is, what is the return?
1:02:34
Speaker 1: Is that stuff better found in the private sector or
1:02:36
Speaker 1: the public space, given your experience?
1:02:39
Speaker 2: So I actually think that's why it needs to be
1:02:42
Speaker 2: a partnership, because public, i.e. traditionally government, is designed to
1:02:48
Speaker 2: deal with the things that don't have a quote-unquote financial
1:02:52
Speaker 2: return on investment right but the private allows you to
1:02:56
Speaker 2: do it more effectively and efficiently so you're not gonna
1:03:00
Speaker 2: you know wasteful spending a lot of times government is
1:03:02
Speaker 2: known for wasteful spending that partnership allows you to support
1:03:07
Speaker 2: this is for the public good and then the private
1:03:10
Speaker 2: is is but we have to do it as cheap
1:03:13
Speaker 2: as possible yeah i.
1:03:15
Speaker 1: Guess the thing that i disappointed in is and you
1:03:19
Speaker 1: know you probably know this better than most which is
1:03:21
Speaker 1: We don't know the discoveries that we'll end up making
1:03:25
Speaker 1: in order to get to someplace that might actually benefit us.
1:03:29
Speaker 1: So what it took to get to the moon led
1:03:31
Speaker 1: to all sorts of important discoveries that helped society here
1:03:35
Speaker 1: that had nothing to do with us being on the moon, right?
1:03:38
Speaker 2: It did. I'm going to circle back to this is
1:03:41
Speaker 2: also why universities are still important. So if you think about-
1:03:46
Speaker 2: We could land the plane. If you think about all
1:03:49
Speaker 2: of the things that came out, not just in terms
1:03:51
Speaker 2: of exploring the moon, but with all of our healthcare discoveries,
1:03:56
Speaker 2: it was professors or group of professors working on something
1:04:00
Speaker 2: where they worked on it their entire life. The Nobel
1:04:03
Speaker 2: Prize in X worked on it their entire life. I mean,
1:04:08
Speaker 2: that's a long time to work on something and there
1:04:10
Speaker 2: is no financial benefit because you're curious and you know
1:04:13
Speaker 2: you can solve it in just one more decade, just
1:04:16
Speaker 2: one more decade. And then you have this breakthrough that
1:04:21
Speaker 2: really changes the world.
1:04:24
Speaker 1: Look, that's why the world is so loves our university
1:04:27
Speaker 1: system compared to everybody else, right? We've been better at
1:04:31
Speaker 1: incubating this stuff than most. And I'd hate to see
1:04:34
Speaker 1: us lose that edge.
1:04:36
Speaker 2: And we still need to fund. That's the thesis of that. Yes.
1:04:40
Speaker 1: Well, the book's Rebooting the Machines. I mean, boy, you
1:04:45
Speaker 1: have a foot in two interesting camps, the engineering side,
1:04:49
Speaker 1: the academia side, this sort of nexus of the two. Boy,
1:04:56
Speaker 1: your observations, I think, are ones that a lot of
1:04:59
Speaker 1: people should be interested in. Appreciate having you on. Thank you.
1:05:03
Speaker 2: Thank you.