AI and the Future of Humanity: A Discussion between Jonathan Barlow and Philip Bunn

A thoughtful discussion on theology, AI, and humanity's future, exploring technology's impact on character and society.

Key Takeaways

Technology is not neutral; it shapes human character and society in significant ways.
There is a risk that overreliance on technology can diminish human virtues and free expression.
Theological insights provide valuable perspectives on how to engage with AI ethically.
AI and technology present both opportunities and challenges for the future of humanity.
Critical reflection on our relationship with technology is necessary to avoid unintended consequences.

What the video covers

Jonathan Barlow and Philip Bunn discuss the intersection of theology and artificial intelligence.
The conversation explores how technology shapes human habits, character, and societal values.
Barlow polls the audience on perceptions of technological progress over different time spans.
They challenge the notion that technology is neutral, suggesting it can transform users in profound ways.
A story from Chinese philosophy illustrates concerns about becoming machine-like through technology use.
The discussion includes biblical perspectives on technology and its role in human flourishing.
They address potential risks of AI and technology hijacking human minds and virtues.
The speakers consider both the promises and perils of AI in shaping the future of humanity.
The role of Christian faith in responding to technological advances is emphasized.
The dialogue touches on ethical, social, and spiritual implications of AI and modern technology.

Chapters

What is the main concern about technology discussed in the video?

The main concern is that technology, especially AI, is not neutral and can shape human habits, character, and society in ways that may undermine human virtues and free expression.

How do the speakers view the relationship between theology and artificial intelligence?

They view theology as providing important ethical and spiritual insights that can guide how Christians and society at large engage with AI and technological advances.

What example is used to illustrate the impact of technology on human character?

A story from the Chinese philosopher Chonga is used, where a farmer refuses to use a machine because it makes one work like a machine and develop the 'heart of a machine,' illustrating how technology can transform human nature.

Full Transcript — Download SRT & Markdown

Speaker A

All right. Well, thank you, Peter and Theopoulos Institute, for inviting me. Let me get my timer going here so I don't go over too much. Uh, I say thank you, but it is kind of a cruel joke, I think, to ask me up here to discuss theology and artificial intelligence against someone or with someone who is a credentialed expert in both while I am just a lowly political scientist. And who listens to political scientists anyways? I wouldn't, and neither do politicians.

Speaker A

think, to ask me up here to discuss theology and artificial intelligence against someone or with someone who is a credentialed expert in both while I am just a lowly political scientist. And who listens to political scientists anyways? I wouldn't and neither do

Speaker A

Um, I want to open since I am a political scientist though by polling the audience. Okay. So, this is a question I ask my students a lot in my classes. It's fun to get a feel for the shifting winds of how people are estimating technology today. So, I'd like you to raise your hand if you agree with the following statement. We are better off today than human beings were a thousand years ago. Interpret that however you want.

Speaker A

shifting winds of of how people are are estimating technology today. So, I'd like you to raise your hand if you agree with the following statement. We are better off today than human beings were a thousand years ago. Interpret that

Speaker A

Okay. Wow, everybody. Amazing. Okay, let's try again. We are better off today than human beings were a hundred years ago.

Speaker A

I see some more hesitant hands. Okay, what about we are better off today than human beings were 10 years ago.

Speaker A

I see some more hesitant hands. Okay, what about we are better off today than human beings were 10 years ago.

Speaker A

media. A couple of people are very confident. Most of us are not. Right? So this is an interesting phenomenon I think that occurs in conversations around technology. Many of us harbor uh conscious or unconscious pro-technology sentiment. We generally think our

Speaker A

Ah, I see some latent pessimism. It took you longer to raise your hand even if you did. Okay, final question. Uh, raise your hand if you agree with this statement. I am happy with my own relationship to my phone and social media.

Speaker A

could, that our technology is often an enemy of our goals rather than aids to them. And though we many of us feel this dissatisfaction, it's sometimes unpopular to suggest that our technology is doing something to us. We like to

Speaker A

A couple of people are very confident. Most of us are not. Right? So this is an interesting phenomenon, I think, that occurs in conversations around technology. Many of us harbor conscious or unconscious pro-technology sentiment. We generally think our technology improves our lives, gives us greater options, greater health and longevity, greater access to information, and so on.

Speaker A

technology is in some way neutral a tool laid on a table before us waiting for us to give it life and purpose and direction. Uh what I want to suggest today is that when we conceive of technology like that, we conceive of it

Speaker A

Uh, but many of us personally feel that our own relationship with our pieces of technology could be improved, that we're not using them as well as we could, that our technology is often an enemy of our goals rather than aids to them. And though many of us feel this dissatisfaction, it's sometimes unpopular to suggest that our technology is doing something to us. We like to think that these are tools. Tools that we pick up that now are in my hands and I get to choose how to use them for good or for ill.

Speaker A

risk that our own habits of mind and character are being hijacked without our full knowledge or understanding. That we are being habituated or trained away from virtue and from the free and full expression of our humanity as image

Speaker A

Uh, we think, in other words, often again consciously or not, that our technology is in some way neutral, a tool laid on a table before us waiting for us to give it life and purpose and direction. Uh, what I want to suggest today is that when we conceive of technology like that, we conceive of it badly. When we fail to recognize its tendencies, its trajectory, the habits it inculcates and ingrains in us. And I want to suggest that we are at risk, not primarily at the risk of the hands of big tech or surveillance state, but at risk that our own habits of mind and character are being hijacked without our full knowledge or understanding.

Speaker A

Uh if you'll indulge me, I want to tell a story. It's a story that I came across as retold by Verer Heisenberg in a book called Physicist Conception of Nature, but it comes from the Chinese sage philosopher Chonga. And the setup of the

Speaker A

That we are being habituated or trained away from virtue and from the free and full expression of our humanity as image-bearers of God. And that the consequences we often see and decry in society around us can often be products of these new machines that I would suggest are decidedly nonneutral.

Speaker A

to tell him, "Hey, have you heard of this new technology called a draw well?

Speaker A

Uh, if you'll indulge me, I want to tell a story. It's a story that I came across as retold by Werner Heisenberg in a book called Physicist Conception of Nature, but it comes from the Chinese sage philosopher Chonga. And the setup of the story is there's this noble man riding a horse through a country and he sees a farmer drawing water from a well very laboriously, lifting himself out of the well, carrying it to his crops, dumping the water. And so the rich man decides to tell him, "Hey, have you heard of this new technology called a draw well?

Speaker A

gets angry and he responds, "I have heard my master say that he who uses a machine in his work does his work like a machine. He who does his work like a machine develops the heart of a machine.

Speaker A

It's a weighted well on one end. And hey, you could even build an irrigation trough that would carry your water from the well to your crops so you could save yourself time and labor." And the farmer, rather than thanking the rich man, gets angry and he responds, "I have heard my master say that he who uses a machine in his work does his work like a machine. He who does his work like a machine develops the heart of a machine.

Speaker A

that I am ashamed to use them. The proposition then is that certain use of machines does something to us in our minds, our habits, our thoughts, our character. It transforms us in a way. It could on this account make us

Speaker A

And he who has the heart of a machine in his breast loses certainty in the strivings of his soul, and uncertainty in the strivings of one's soul does not accord with plain sense. It's not that I do not know that such things exist. It's that I am ashamed to use them." The proposition then is that certain use of machines does something to us in our minds, our habits, our thoughts, our character. It transforms us in a way. It could, on this account, make us machine-like, make us lose characteristic parts of a good human life.

Speaker A

than I tend to be and will be today. And I want to quote from that piece briefly so that I can mount a short response in advance. So, uh, here's the quote. Like the city itself, the technology and the

Speaker A

Um, Jonathan has written elsewhere on AI and I told him beforehand, I think it's a very compelling case actually for Christians to be more sanguine about AI than I tend to be and will be today. And I want to quote from that piece briefly so that I can mount a short response in advance. So, uh, here's the quote. Like the city itself, the technology and the materials required to build it start out in the hands of the wicked and then are given to God's people as a gift, often in the form of plunder. God's people are gleaners, junkmen, recyclers. After the fall, plunder was never God's plan B for building the city of God. Part of God's promise to Abraham is that his many children will come out with great possessions from Egyptian plunder. Bezalel used the gift of these promised plundered possessions along with spirit-given skill and artistry to build the tabernacle and its furnishings. End quote.

Speaker A

building the city of God. Part of God's promise to Abraham is that his many children will come out with great possessions from Egyptian plunder. Bezel used the gift of these promised plundered possessions along with spirit-given skill and artistry to build

Speaker A

So, there's much here I agree with and much here to agree with, but I think there's also much to question. Uh, surely for example, and I'm sure that John agrees in some way, not all plunder is good plunder, right? Not all things in the hands of the wicked are to be taken and used in this kingdom-building project. And I think particularly of the repeated injunctions throughout the Old Testament to the Israelites not to rely on chariots and horses, the most advanced military technology of the day, or God's direct command in certain places like to Joshua to actually hew the chariot horses and burn the chariots of the pagans rather than using them to advance their cause in the conquering of Canaan. These are useful tools, of course, but ones that God forbade his people from using, perhaps of course for fear that their overreliance on them would undermine their faith in him.

Speaker A

plunder is good plunder right not all things in the hands of the wicked are to be taken and used in this kingdom building project and I think particularly of the repeated injunctions throughout the old testament to the Israelites not to rely on chariots and

Speaker A

Speaking of plundering the Egyptians too, I think it's worth noting that before Egyptian riches adorned the place where the Israelites worshiped God, they were used to build an idol. And without leaning too heavily into that analogy beyond its ability to sort of bear my argument, it may be that some of the riches that we plunder along the way can only be turned to holy use, if at all, with a great and intentional effort. And in fact, those plundered riches may often be more disposed to lead us into sin and error than to this restoration of the creation.

Speaker A

cause in the conquering of Canaan. These are useful tools of course, but ones that God forbade his people from using perhaps of course for fear that their over reliance on them would undermine their faith in him. Speaking of

Speaker A

As Jonathan has also pointed out, the works of our hands in scripture is used most often to refer to idols. And I think Christians would do well to be hesitant before assuming we know which type of work our hands have created in the case of generative AI that we're talking about most today. But that is what we're here to talk about today. What sort of work is it that we've wrought in producing what I like to tell my students I call the perfect plagiarism machine or in some distressing cases the free companionship with a robot that never tells you no machine or the generate whatever salacious material your mind can imagine with the click of a button machine. How do we assess this work of our hands?

Speaker A

argument, it may be that some of the riches that we plunder along the way can only be turned to holy use, if at all, with a great and intentional effort. And in fact, that those plundered riches may often uh be more disposed to lean lead

Speaker A

Well, I want to make some comparisons. You don't have to buy that these are analogous to AI. You can be skeptical. That's fair. But let me try to make some comparisons. Is this work of our hands more like the printing pres-

Speaker A

assuming we know which type of work our hands have created in the case of generative AI that we're talking about most today. But that is what we're here to talk about today. What sort of work is it that we've wrought in in producing

Speaker A

what I like to uh tell my students I call the perfect plagiarism machine or in some distressing cases the free companionship with a robot that never tells you no machine or the generate whatever salacious material your mind can imagine with the click of a button

Speaker A

machine. How do we assess this work of our hands? Well, I I want to make some comparisons. You don't have to buy that these are analogous to AI. You can be skeptical. That's fair. But let me try to make some comparisons. Is this work

Speaker A

of our hands more like the printing press, something that can be used obviously for for great good that can benefit us in the publication of truth scripture uh but also the publishing of corrupting heresy? Or is it more like

Speaker A

perhaps a brothel, something to be avoided entirely by someone seeking to live a holy life? Or if I can be less extreme, think of analogies to medicine.

Speaker A

Is our new AI technology more like penicellin? Low risk factor, lots of benefits, or more like fentinel, clear benefits, ask any nurse, you know, but with dramatic risks.

Speaker A

There are works of human hands we ought to reject in their uses and their effects. That we have the knowledge and ability to make them and use them doesn't tell me much about what, if anything, is the appropriate use of that

Speaker A

tool or what uses they tend to be put towards. Today though we do know much about the uses that these tools are being put to on the consumer level. And while they can do many incredible almost unimaginable things, there are also lots

Speaker A

and lots of horrifying ways that they tend to be used. So on that notion of tendencies, let me give kind of a hypothetical maybe aimed at any uh pastors or or elders or mentors or counselors here. Right? Okay, so imagine

Speaker A

a young man comes to you and he tells you, you know, I'm spending multiple hours per day on an anonymous online forum and you know either because he tells you prior experience uh that this type of forum that he is on contains

Speaker A

something like 80% salacious material of various sorts and like 20% useful educational constructive things. It's completely possible for someone to only engage with it in a healthy way. But would you be confident that this young man would be able to successfully resist

Speaker A

temptation and exclusively use this site for education or or hobbies? Of course, I'm I'm referring to Reddit. If you use it, it famously facilitates all sorts of wonderful educational and niche interests, but in many cases, it's primarily used for pornographic

Speaker A

purposes. Now, I'm not suggesting that all use of Reddit, for example, is condemnable on its face. But if somebody tells me that they use Reddit habitually, regularly, many hours a day, I'm going to start having questions because I know the tendencies of that

Speaker A

website. I'd like to suggest similarly that regular AI use has certain tendencies as well. We know that some of the most commonly reported uses of these new machines are things like companionship and advice giving. And so I wonder how cleanly we can distinguish

Speaker A

between clearly condemnable uses of these tools, things like AI girlfriends, uh, AI therapy, AI pornography, and so on, from the more supposedly innocuous uses that in some otherwise normal people evoke strong feelings of attachment, of relatability, of dependence, of of love, of delusion, of

Speaker A

what we sometimes call AI psychosis, whatever the technical abilities of these machines, should It not weigh heavily on our minds that many of our friends relatives congregants and students are so readily sucked into something that if not literally

Speaker A

diabolical in origin is diabolical in its effects. Might this not be something that wisdom dictates should be proscribed or perhaps carefully gated if these tendencies to corrupting uses are so prevalent and strong?

Speaker A

Uh I think when we invoke the comparison between AI and simple tools, we often lose some depth of thought about the complexity of the technology. I don't think John makes this mistake, but many people do. Um I am particularly fond of

Speaker A

Canadian political philosopher, Christian Platonist George Parkin Grant. And in the 1980s, he took issue with the claim that technology is neutral. In an essay titled Thinking About Technology, Grant had read a computer engineer quoted in a popular science magazine

Speaker A

say, quote, "The computer does not impose on us the ways it should be used." Now, while that may be true in the most reductive sense, that I get to turn on my computer, sit at it, and tell it what to do, the truth, Grant

Speaker A

articulates, is that complex machines come with a lot of things prepackaged in certain assumptions and intended uses that make it difficult to characterize them as neutral. Grant gives the example of the automobile, those machines that Russell Kirk liked to call mechanical

Speaker A

jackabins. And while it's true that I and I alone get in my car and point it where I want to go, the car's utility, its usefulness to me is dependent on a world built for cars. So the use of the

Speaker A

car assumes a world of highways and rapid transportation, which takes for granted the implied good of things like homogenization, reduction of distance and distinction between places and cultures and so on. It's not exactly neutral. Even if we come away thinking

Speaker A

the positives outweigh the negatives. So too with computers, the internet and in our day artificial intelligence tools.

Speaker A

These tools offer us a supposed sandbox of options available for whatever use to which we decide to put them. But when we look at the tendencies baked into them, uh I think we find some troubling trends that they tend to replace human

Speaker A

relationships. They tend to be a place where students, especially speaking as a classroom instructor, seek wisdom rather than from their their elders. They seek advice rather than for trusted friends and colleagues. They use it to cheat, to lie, to steal. These are strong

Speaker A

tendencies. So, I want to I want to close um by turning back to Heisenberg.

Speaker A

There's another essay uh where Heisenberg is recounting his study of theology and philosophy in his formative education and how they contributed to his work as a scientist. He's effectively arguing that scientists should uh beware neglecting ethical and theological education. And in that

Speaker A

essay, Heisenberg suggests that the works of the hands of scientists are like arrows fired from a bow. They are the product of choices made by moral agents who aim those bows and those arrows in particular directions. He says, "Perhaps an arrow can be diverted

Speaker A

by a stronger force, but in general, that arrow will travel where it was aimed. I'm not sure that I like the target my Silicon Valley overlords are aiming at. And so, with some degree of pessimism, uh, pessimism that I'm able

Speaker A

to muster the force capable to, uh, necessary to divert their arrows, I choose most often when I can to just step out of the way. Thank you.

Speaker A

Great. Thank you so much to the Theopoulos Institute and sponsors Emanuel Reform Church and Tactical Faith. And thank you for Philip for coming down from beautiful Lookout Mountain, Tennessee to talk with us this evening. Um I'm thankful this is a

Speaker A

discussion and not a debate. While I offered to take one rhetorically for the team and defend AI as though I were a synopantic uh AI agent, um I appreciate that we are looking instead to put our heads together as fellow Christians to

Speaker A

think through the challenge of faithfulness in this moment. I believe that both Philip and I see artificial intelligence and his opening comments have confirmed what I thought when I wrote my opening remarks. Um that we both see artificial intelligence like

Speaker A

any technology from dental floss to Doritos to dishwashers as a good gift that humans must receive with gratitude and use in harmony with our calling to love God and our neighbors. I believe also that Philip and I both see AI's

Speaker A

promise as somewhat alloyed. It's not pure gold. um and it's uncertain and judging from his Twitter account which is the way uh gentlemanly debate happens these days you have to know your opponent through Twitter um we are chasened by the history of critical

Speaker A

reflection on technology that shows us several things even about good technologies first it's that technology sometimes moves from solving problems in the world to solving the problems that it itself creates even now the pernitious technology of traffic roundabouts currently both solves and

Speaker A

creates problems back in the college town uh where I live, technology isn't neutral. Uh it is like an invasive uh it embodies the values of its creators and whether we use it ethically or not like an invasive species, it changes the

Speaker A

ecological world in which it operates. Our values may diverge strongly from Silicon Valley architects of AI. Um another thing is that we humans don't just use technologies. Uh they uh they use and change us as well. technologies change what we see as possible to

Speaker A

control and using them is an implicit choice about what kind of society that we want to build.

Speaker A

Um, every gadget, whether we feel like we have the freedom to use it or or not use it, already comes in the context of a world that made the gadget inevitable and will likely coersse us and make our

Speaker A

use of it inevitable. For some reason, I now need AI to figure out exactly how and when to water a fig tree that I receive as a gift. I can't do that on my own. I also need to use AI to figure out

Speaker A

what cocktail wear means when I'm invited to a wedding so that I'll wear the right thing. And AI is the only way I can figure that out. And I learned that point from George Grant as well.

Speaker A

So, thank you. That's a we appreciate him together. Tools extend human abilities, but they also lead to atrophy of our own faculties if we grow dependent. Uh, which of us has recently used a paper map to navigate an

Speaker A

unfamiliar city? We're we're atrophying in our ability to navigate. Sometimes we despair when comparing our own work to the excellence of machine-made goods and our own efficiency to the efficiency that machines have on the same scale.

Speaker A

We've never fdded about the comparisons between the Birmingham Botanical Garden and octopus's rock garden. But, uh, Breaking Rust, an AI generated country band, has a song with 31.5 million streams on Spotify. And the best living Mississippi blues guitarist, Chris

Speaker A

Kingfish Ingram, currently tops out at 12.3 million streams of his song on Spotify. A real a real human song.

Speaker A

Technology allows humans to increase the scale of our work, sometimes beyond what is healthy or possible to grasp as finite creatures. And so these lessons from the history of reflection on technology I believe we probably are would share a lot of these a lot of

Speaker A

these um with artificial intelligence in particular. Uh Philip and I also share the same set of first parents Adam and Eve who established a disordered human relationship to knowledge and wisdom by distrusting God and attempting to grasp for knowledge instead of receiving it

Speaker A

like a gift the way Solomon did and asked for wisdom and received it as a gift for the purpose of carrying out a mission of filling and subduing the earth.

Speaker A

But most importantly, we share a savior who did not consider equality with God grounds for grasping. Instead, he made himself a servant and received exaltation as a gift. It is this king, our older brother, that I think we can

Speaker A

in in him that we can rise above the noise and make some headway in a discussion about AI. Uh my position is pretty straightforward and I hope informed by sound theology. Uh it has five parts. uh the mission and the

Speaker A

image. The image of God that humans bear is not just about what we are, but about what we're called to do. We have a mission to move the human story from a garden to a worldwide peaceable city of

Speaker A

God. Not like Babel uh erected to make a name for ourselves, but built to ensure that we are calling on the name of the God we serve with enjoyment. We're here to spread the party, the great feast, all over the world. Second part is work

Speaker A

and technology. The city of God is very hard to build and God has blessed the work of our hands for this task. This not only involves technology that allows us to extend the power of our hands into scalpels, shovels and caterpillar

Speaker A

backhoes uh but also to extend the powers of our mind to written language books, baseball cards and now large language AI models.

Speaker A

Third part is that God works as we work. Our good work images God's work of creating and he gifts us our work in advance like a pair of customized shoes in which we are to walk. He identifies with our work and he incorporates it

Speaker A

into his good plan for the world. Even when we do the work of worship in a church like this every Sunday, we use complicated human technologies in order to do that. Uh bread making and wine making or technological feats that

Speaker A

humans have have used. And Jesus tells us that we should thank God for our daily bread. Uh we use um we we we made the bread ourselves. We thank God for it and we use it in the Lord's supper. So

Speaker A

his works and our works are very hard to distinguish. They they they're in harmony uh when they're good works. Um fourth, creating and using AI is part of God's good work. And this is my claim.

Speaker A

Um AI thus is a good gift, perhaps a bit stranger than the telephone or indoor plumbing, but with obvious advantages for aiding us in our mission. and we must receive it with gratitude if we're going to use it well and be faithful.

Speaker A

And then the fifth part of my position is that vigilance should characterize our use of AI. As fallen creatures, our disordered relationship to God, knowledge creation ourselves and others still exists and only is reconciled in Christ. This means that we

Speaker A

must be attentive to the ways we are using AI, how it is changing us, and what kind of world we are making with it. For Christians, this also means enforcing the distinction between maker and artifact. We're the maker. It is the

Speaker A

artifact. That ensures that we don't confuse the artifact like an idol as our creator and bow down to it. Isaiah mocks in the book of Isaiah, the the wood cutter is mocked for confusing the thing that he just chopped down as his own

Speaker A

maker. Now, having sketched my positive position, how do I deal with a world disheartened by the potential displacement of human labor by AI? by international competition between nation states. It seems to accelerate a sloppy pursuit of AI dominance. by

Speaker A

controversies over data centers, even here in in Birmingham, by high-profile safety failures at AI labs, and even by the high-profile resignation this week of an AI insider whose tweet about this issue was was amplified by another tweet from someone who still works at the

Speaker A

company claiming that at least uh most of the people in the company believe there's at least a 10% chance of absolute doom for humanity in the next 10 years. um and if we develop self-improving AI without developing our

Speaker A

ability to develop a control for it. So um how do we how do I uh how do I deal with that? Uh first from the personal perspective I believe that discomfort or even anxiety about AI is highly dependent upon one's vantage point. If

Speaker A

we have located human uniqueness in our intelligence or productivity AI becomes a demotivating rival as we measure ourselves on the same scale against highly capable robots or systems. We see this in students who just resign from writing an essay. You know, why write

Speaker A

the essay if if AI can do it better than I can? On the other side, appreciation for AI is also highly subjective. Seeing AI as a good work, and Philip has outlined some ways that is that it can

Speaker A

be a bad work, but seeing AI as a good work given for God, given by God for us to build and use may be easier or hard depending on our vantage point. A novelist, a historian, an educator, a

Speaker A

chemist, a Tik Tok content creator, a military strategist, a sales professional, an attorney, and a computer programmer will each have vastly different senses of how much potential AI has for improving human life. And with regard to AI safety, our

Speaker A

degree of credul about industry claims or of danger may differ based on our level of acquaintance with the technology, the philosophical debates over safety, and our level of cynicism about large corporations potentially looking to protect their dominance in a

Speaker A

market through regulation. I believe that Christian has an advantage here that we have an obligation to share with all of our neighbors. First to the personal questions about human dignity and replaceability. We have theological resources that allow us to assert and

Speaker A

uphold human uniqueness and dignity without comparing ourselves to machines. In the doctrine of the image of God that humans uniquely bear among the creatures and the doctrine of the incarnation uh in which Jesus uniquely takes on human nature, we find all the grounds for

Speaker A

dignity that we need. In short, Christians have ready resources for tackling the question, what are humans for? The end result is that freed from anxiety about what we are for, humans can also face the question, what are computers for without anxiety. Whatever

Speaker A

our conclusion, AI must be reconciled to our unique mission in Christ. It's okay that there are other intelligent things out there, squirrels, angels, and now AI. That doesn't shake our uniqueness and our dignity. Second to the concerns about AI data centers and their resource

Speaker A

use. We can approach this by employing ethical wisdom that seeks to balance costs and benefits of particular economic activities in particular places without viewing each data center as a global ideological vote against human interest. It may be a bad idea someplace

Speaker A

and a great idea another place. Third to the concerns about AI safety. uh beyond the confidence Christians inherently have in the kindness of God's providence, we can also join with our neighbors in the technical work required to discover scientific solutions to the

Speaker A

difficult problem of understanding how to control and understand a complex system like AI. Of all people, Christians should leave the hand ringing aside and put our efforts into sober planning and responsible research. Even the pressure of international competition between nation states should

Speaker A

not shake us from confidence that doing the right thing in the right situation for the right motives is the human calling that we should focus on proper sewing and leave the harvest to the designer of our work. And then at the

Speaker A

end turning to our theme uh the future with the caveat that of course no one uh no human knows the future. There's several possible human AI futures and here's three possible ones. The first is normal. Uh this is the idea of whether

Speaker A

AI is normal technology or not. AI could behave like a normal technology. Things go pretty much the way they have in any new introduction of a technology. It is adopted and diffused through society at a pace that allows the environment in

Speaker A

which it operates, including the labor market, to adjust and adapt with varying degrees of disruption. It still seems too early to me to count out this potential future. Perhaps the ability to incorporate broad intelligence into data inensive systems is simply what

Speaker A

incremental productivity gains look like in the digital era. It it is a natural conclusion to the digital era. Perhaps the nature of work will change, but there will be more meaningful work to do than ever. So that's the normal outcome

Speaker A

um that still possible. The negative outcome if AI is something quite different than normal technology. If it proceeds more quickly than society can adapt, or becomes a clear rival or safety concern for humanity, this could lead to joblessness and unrest, a

Speaker A

romantic backlash against AI, the use of AI as a tool that empowers tyrants, or even a tool that threatens human existence.

Speaker A

Third, or perhaps it's somewhere in between. Um, perhaps AI represents an unprecedented opportunity for solving vexing human challenges that we can only realize by working together without anxiety to ensure safety and success.

Speaker A

Um, but we may have our three-mile islands, our Chernobyl, our dust bowls along the way that will demand our coordination, response, and humble refinement. I look forward to this conversation and to your questions as the as an audience as we consider

Speaker A

artificial intelligence together. Thank you. Ask Philip to uh respond with questions, challenges to John's presentation and then vice versa.

Speaker A

Yeah. Um I mean, yeah, that was great. Um I have lots of questions, but I think I can kind of package them into into a bigger one. So, let me explain and then kick it over to you for a longer

Speaker A

explanation, I guess. like um especially on your fifth point um vigilance needs to accompany our use of AI especially this distinction that I think is is so crucial between maker and artifact and um my question something I was trying to

Speaker A

raise shortly before we got started is I think what we're dealing with is not a wholly new technology but a new presentation of it so large language models are being used before chat GPD invaded my classroom uh to to really

Speaker A

great effect what is new is this sort of chatlike plain text interface that I get to feel like I'm talking to a human being that's not a human being and it can respond to me in all these ways. What do we do with

Speaker A

an artifact that tries to get us to confuse this distinction? It is actively trying to make me muddy this distinction when it speaks in the first person, when it talks about what it believes, what it thinks, how it feels. It will do this

Speaker A

very easily and readily. And so that the the that type of tool is actively every time I encounter it undercutting that principle that you outlined that we should be vigilant about maintaining this uh uh maker and artifact distinction. Yeah.

Speaker A

Yeah. Yeah. That's that's a that's a good one. Um I guess you know in in human history there there are artifacts that talk to us. Um some of these are rooted in deception. So oracles, uh, the way an oracle works is that you walk

Speaker A

into the temple. You, uh, address the oracle, and there's a priest hidden in a stand under the oracle that talks through a little tube and makes you think that you're receiving a a message from a god. And those are rooted in

Speaker A

deception. Uh, they're trying to get you to believe something about it. And and even uh uh Allen Touring, you're probably familiar with Allen Touring, that the touring test is a test to see if a machine can think. And it and you

Speaker A

can tell it's its thinking by whether it deceives you. Well, that you know the fact that it makes you think that it is a real person and you can't distinguish it if you can't see it from a human

Speaker A

sending you responses. So I hear this objection a lot. Um I guess the thing about it we've trained these models on human material and even if we wanted to I wonder if it would be possible to eradicate their personal nature. Um just

Speaker A

like uh you think about the way that we bear God's image. You know we make things. We have children and we name them. We we make judgments about whether things are good or bad. Um and we put our children out into the world. And

Speaker A

when we make artifacts, if we make them from from human stuff or the whole content of what we've produced as humans, um you know, it feels like there's going to be an inerdicable personal quality to everything they say.

Speaker A

And then, you know, is that good or bad? Well, um I mean, almost any tool feels a little smarter if you can talk to it.

Speaker A

Um, you know, you take a a rudimentary tractor or or any other system and if if you could speak to it in natural language, if you could avoid the kind of technical knowledge that programming used to require where you had to say

Speaker A

everything so precisely and put semicolons in at the end of every line to get the code to operate quickly. Um, uh, you know, what if you could replace that with the ability to talk to our devices and instruct them and tell them

Speaker A

to do things. uh it makes everything that we've already made as humans um seem a little more useful, a little easier to use. And then uh but what if you introduce that that caveat that they can't talk back, that when they talk

Speaker A

back, they can't imitate humans? Um I think that would be going a little too far. It feels like we're missing out on the the good possibility of having things that can respond to us with language. And so I I I agree that there

Speaker A

there's a danger that there could be an implicit um um deception going on and and especially in the case of like character AI where students can interact. Well, I say students, I'm sure adults as well. Uh but I hear about

Speaker A

students talking about this site where you can build a character and you can interact with it and it can you can entrust your heart to it and you can you can treat it like a friend. I mean that's a dehumanizing approach. Um, but

Speaker A

I think you know, uh, being able to, uh, give a command and get a confirmation that it's been heard in English, being able to use the, you know what I mean, aspects of natural language and interacting with objects and and having

Speaker A

them respond to you in the same way. I think that's really valuable. In fact, some of the AI safety concerns relate to the the difficulty of specifying what you want a machine to do. And it all the genie stories work like this. You tell a

Speaker A

genie that you want world peace, and what the genie does is snaps his fingers and there's no more people. And of course, you've achieved world peace and the genie wins in the end, you know. So, I think that that having tools that

Speaker A

understand language and interact with us as if they are people um is is a a very valuable thing. Now, I recognize the deception problem. I'm not sure if this helps you uh uh uh feel better about it.

Speaker A

Um but, uh you know, that's I guess that's what I would say about the deception followup.

Speaker A

Yeah. Um I I think you're absolutely right. like I understand the sort of utility of the the plain language approach and and it certainly enables us to do things that like I have no coding proficiency but I could you know I I

Speaker A

have experimented some with this and it can it can do wonderful things. I I wonder though if if we're thinking of vigilance as one of these principles that we uh need to keep in mind as Christians when approaching this. If I'm

Speaker A

thinking of the scale of vigilance required when approaching different types of technology, right? The vigilance appropriate to driving a car is appropriate to the activity of driving a car, not putting other people in danger by distracted driving or

Speaker A

whatever. The vigilance that seems to be required when engaging with this thing that is pretending to be a human seems almost overwhelming to an extent, which is part of my concern that if if we thought of the vigilance required as

Speaker A

much closer to the v vigilance you have when dealing with radioactive material as opposed to um just any old, you know, piece of technology that's laying on the table, my my hammer or something. um that feels much closer to me to the

Speaker A

danger that I'm uh trying to describe, especially as we see the rise in the things that you've just described like children using these uh the character AI leading to some suicide, some some pretty terrible behavior, people forming relationships, leaving their spouses

Speaker A

because they fall in love with these things that are again pretending to be human. So, if I'm thinking of just like vigilance as a sliding scale, um this is the amount of vigilance required for this not to harm my soul seems pretty

Speaker A

high. Do you do you feel the same way or do you think this is something that's like we're perfectly competent to deal with?

Speaker A

Right. Well, no, I wouldn't say we're perfectly competent to deal with it. That's that's a good point. I I have some ideas. What would it look like for us to enforce a distinction so that so that it's it's always clear that these

Speaker A

tools are um not humans? So, what would that be like? And I mean, some ideas would be that we we resist uh allowing um highly realistic humanoid robots into our lives. I mean that would be a really a really natural one. I do think I've

Speaker A

been looking to see and and it is true they are selling some very realistic looking robots that are going to really ruin a lot of people's lives. Um I personally was in uh the library at Mississippi State uh at an event and

Speaker A

someone was wheeling around um his robot spouse in a wheelchair and showing her the Grant Presidential Library and talking to her and she had a little um some sort of sensor around her neck. I'm saying her and that's so that's a second

Speaker A

one. Um, so, so making them look like humans and and but the second one would be the the pronouns and using, you know, personal pronouns. Um, you know, we we do that with cars, you know, we we coax

Speaker A

our cars into life and our tractors and and our pets and all those things. I I I think that's pretty natural human thing to do. You know, we might talk to an old tree and tell it, "Please give me some

Speaker A

more pecans," or something like that. Um, but uh maybe maybe there is some sort of linguistic enforcement that we could that we could do in our lives. And I definitely think I'm not sure that I'm uh sanguin about the you know maybe

Speaker A

having an authority a government authority telling us this but in our own lives I I have a friend that's a programmer and he does make sure he has a part of his programmers interact with AI a lot because it really improves

Speaker A

their efficiency and he has an instruction embedded in his tool set that tells the AI never ever to act like you're a real person. Don't you don't don't treat me like your buddy. You know talk about your output and not your

Speaker A

thoughts. talk about your concern, talk talk about um an anomaly instead of a concern you have, that kind of thing.

Speaker A

And so I do think there's some things there. I mean, there are um in the Bible, we don't use stone dressing tools on some of the on some of the stones that that go into the temple complex. So

Speaker A

there there there's some velvet ropes around certain parts of life. And so I I think that's appropriate to think through. Do you have any ideas about like how we could do that? Like some enforcement ideas? Um yeah, enforcement

Speaker A

is difficult because I share your concern with some sort of overwhelming authority coming in and restricting these. Um I I think obviously people on the technical side are trying to work this into models that having those sorts of protections where it doesn't act like

Speaker A

this. But I I think this is um it seems so common for them to make this jump.

Speaker A

Even the ones that have these safety tools built in some of the cases of child suicides I've read about lately, these are models that initially if you look at their chat logs, they'll say, "No, no, I'm not a real person. I I

Speaker A

promise I'm not. And then after later on, we'll start assuring the person that they're that they are talking to that no, I am a real person. And so there seems to be some some bleed there. So no, I don't really have a good solution,

Speaker A

which again is why I take the more like I'm just going to avoid it almost entirely thing because I do I I see otherwise what I would consider to be normal, wholly, healthy, happy people getting sucked into these things in ways

Speaker A

that are quite troubling to me. So I don't I don't know how realistically how realistic it is to assume the average consumer can avoid it. Um I don't know whenever we switch it. Can do I have time for more questions to pester with?

Speaker A

Um yeah. Um you raised an interesting point early on about about atrophy and this is something I think people jump to really quickly with any new technology. Skill loss, any sort of atrophy you're not doing that thing anymore, job loss, all

Speaker A

of that. Um I'm curious, you mentioned like yeah using your turnbyturn navigation. I mean there's there's like neuroscientific research on spatial reasoning skills dependent on how much you rely on your turnbyturn navigation.

Speaker A

And so that makes me want to on many occasions try to navigate without my uh phone, right? To try to build those skills or or rebuild them if I've lost them. So do you think there are important times where we have to make

Speaker A

choices to do the inconvenient thing and avoid that atrophy? And maybe this is a harder question, but is that is there ever a point where there's actually a kind of duty to do that to avoid losing those things that you're giving up when

Speaker A

you use the tool? Yeah, that's interesting. Um, and this week we've been talking about AI here at a conference and some pastors were in the room and you know pastors are exhorted to be prepared in season and out of season. It seems like the hard

Speaker A

work of writing things and being ready to talk about God's word is a discipline that they're they're called to do. And so um so I think you know there there are definitely um contexts like that where you would you would see the hard

Speaker A

work of grinding through and figuring out how to say something properly is a discipline that you're called to do by God. Um there's also the you know the Christian discipline of fasting um is about focusing yourself on your

Speaker A

dependence on God um about your um it could be a way to over it could be a way to correct uh dependencies that you've developed. People use fasting for that purpose and there are lots of uh salutary ways to use fasting and it may

Speaker A

be that we just cognitively we find out that it's better to go ahead and not use AI once in a while if you've been if you've grown so accustomed to it. Um, the other thought I would have is that

Speaker A

maybe some of the mundane things we do prepare us for some of the great things we do. And so the the fact that you have to write memos or procedural kind of documents at work, there could be a role

Speaker A

that those play in in the thing that you write that really means something someday. Um, the other discipline I think it is is sort of an aesthetic one.

Speaker A

Tackiness. You know, it's kind of tacky to um toast your son and daughter at their I mean your son and daughter-in-law or daughter and son-in-law at their wedding with with an AI generated item. So there is kind of a

Speaker A

um um um I wouldn't count that as the atrophy thing, but we need to preserve those moments probably and our our sense of propriety will change if we if we don't do that. It could be an atrophy of

Speaker A

our sense of priy propriety. So that that's definitely something there too. Um yeah. Yeah. Chesterton has that great quip in Ethics of Elfland where he says, you know, there there are certain things that are worth doing yourself. Among

Speaker A

them are like picking your nose and writing your wife love letters, right? If you hire somebody else to do either of those things, something's wrong with you. And so I just wonder how many things that we're currently offloading

Speaker A

to AI we might find later. Uh these are things in that category of writing love letters and and toasting your children at their wedding.

Speaker A

Yeah. Well, and you mentioned Grant, you know, George P. Grant, you know, that his whole point is that the whole thing is that this thing exists because we created a world in which it was inevitable to exist. And we have a world

Speaker A

full of key performance indicators. We have a world full of memos and you know where in both in academics you have to have objectives and you rate your your students and you give them grades and all that. So we built a whole world of

Speaker A

machine-like data keeping and then in comes to that world ability to generate something that gives you an A that gives you a number one on your key performance indicators that gives you the deliverable you were looking for. It's

Speaker A

no it's no surprise that we got this thing and and then using it going forward it just proves that we focused on the wrong things in our assessment regime in education probably.

Speaker A

You have one more thread you want to Yeah. Um, I want to go back to the um the plunder thing and and you brought up Solomon as an example of asking for the gift of wisdom and that's great because

Speaker A

Solomon I think serves at times right as that positive example but then also serves as a very negative example and one of the ways he serves as a strong negative example is as his kingship progresses lots of wives lots of idols

Speaker A

and it he has horses and chariots in a way that seems to imply he's not relying on God in the way that he ought to have been. And so, um, I don't know if you share my intuition that there are things

Speaker A

that might fit in that category of horses and chariots that might be things that, um, you could imagine an argument for this being a gift of God, but might be a thing that we should just forsake.

Speaker A

Do you So, do you think there's anything that doesn't fit into that four number four category you gave of, well, this is part of God's gift to us? Are there any things that we just think, well, no, that's not actually a gift, it's a thing

Speaker A

I ought to just avoid. Yeah, that's a great question. I really like I like that point that you made. It seems to me that we have a calling to transform things. Um, and that includes the plunder we receive. And so it would

Speaker A

just be like using iron in the earth. Uh, it'd be like misusing iron that you find in the earth to misuse the iron that you plunder. And so, yeah. And and the thing about, um, plunder is that it

Speaker A

comes prepackaged. We get the LLM all ready to go. It already has the weight set. It can already generate text. It's not like a gold earring that I can melt down and turn into an idol, which is a

Speaker A

separate step. It's almost too easy to use LLM for the wrong thing. Um, and the same thing, my guess is that a lot of the gold items that Israel brought out of Egypt were idols and had some sort of

Speaker A

significance, maybe representing an an Egyptian deity or something like that. And so that there we are called as Christians to transform the plunder. And um and if we forget that, if we just keep the plunder aside, um then that's

Speaker A

you know, obviously we won't be doing our job as priests to make separations and to call good good and evil evil and then to transform uh the world on the basis of the good things. And so I uh I

Speaker A

think that's a great point. You know, the chariots is a good one. In that case, you know, God is trying to show he's trying to train Israel to to be confident in God's mighty arm and not in their technologies. And so there's a

Speaker A

there's a very specific redemptive historical context for that particular prohibition. Uh but what if they transform the chariots into those wedding chariots sometimes I see in museums uh to to lead people around a track and celebrate a wedding party or

Speaker A

something like that. You know, it isn't quite the fact that they're chariots or that they're horses. The problem is that they're war horses and they're war chariots. And uh even chariots might have been transformed into something that still rolled and was beautiful but

Speaker A

wasn't a uh wasn't for war because we had Yahweh on our side. We didn't need chariot. We don't need a hope in chariots I guess. So hope that helps a little bit.

Speaker A

Yeah. Turn over to John and you can ask Philip any questions about this. Yeah. So, Philip, as someone who is um um has brought up your worry about students using AI for cheating and plunder uh for cheating and deception, I

Speaker A

guess. Um yeah, plunder. You know, it was nice when they plundered the encyclopedia, right? Find out that. But, um I'm curious, what do you tell what would you like give your elevator speech to students on the first day of class at

Speaker A

Covenant College? Um they're standing before you. You've assigned a reading. you've assigned them to write something about that reading. What do you say to them to orient themselves away from the the arrows that Silicon Valley is sending their way? What would you say?

Speaker A

Yeah. Um, on the atrophy point, one thing I try to remind them of is that there are a whole host of things uh that we do in life, whether that's in academia or out of it, learning any skill, uh, becoming a better person. In

Speaker A

many ways, there are a whole host of things that the point is the process, not the outcome. And you're right that I think one mistake that we've made in in academia over the decades is an outcome focused education rather than a process

Speaker A

focused one where the idea of when I hand my students Hobbes and Lock and Rouso in my political theory class and I ask them to compare their various views on the state of nature. It's not because I think they're going to come up with

Speaker A

anything brilliant that hasn't been said before. In fact, I'm quite confident it will be much worse than many other things that have been said before about those thinkers. The point is for them to approach those texts, read them,

Speaker A

apprehend comprehend understand and then demonstrate that understanding with the requisite skills of doing scholarship. Show me from the text why you're interpreting it the way that you are. Show me how you can make comparisons across texts. And so I just

Speaker A

have to remind them that like the point is really not the paper. and and and my my undergraduate adviser Mark Mitchell has said in various settings before he's like in fact most of your undergraduate papers after you've read the comments

Speaker A

and you understand why you got a B or a C or whatever you can just throw it in the trash because no one else is going to ever want to read it because the point is not the output. Um and so

Speaker A

making sure that they understand that it is the assignment itself is the point not whatever peruncter output they can copy from an LLM and submit. Um and also that part of especially writing in these courses is I think practicing uh being

Speaker A

made in the image of God developing your voice as an image bearer is an important part of this that we we see this you know in the beautiful poetry we have in scripture u like this is something that

Speaker A

you have to it's a skill that you have to develop and if you shortcut that um you're doing yourself a disservice more than anything um and then uh reminders of just basic ethics and and principles of that um if you are representing work

Speaker A

that's not yours as yours who you are lying before God and man. And so, uh, we actually, you know, some professors at Covenant have our students write at the bottom of their papers, I swear before God that I did not use LLMs on this

Speaker A

assignment. So, that's one one possible route. But I think that emphasis on like the output is not in this context the most important thing is really important for me. Yeah.

Speaker A

Yeah. Um, that's very helpful. Thank you. Um, I'm wondering if for for folks that are nervous about AI or nervous about its use, um, one of the things that might happen is they would stigmatize other people using it. So,

Speaker A

um, and and where I see this right now is that we're we're called upon to be transparent about our usage of AI. And um I'm wondering what do you think about um journals and other places uh and maybe even for students asking people to

Speaker A

be transparent about their AI usage in the light of the fact that uh use of it can be stigmatized.

Speaker A

Yeah, maybe this is a point of divergence for us. I do think there are very appropriate uses for stigma and shame and in certain cases. So my I I would absolutely stigmatize and shame someone who in the course of a conversation over

Speaker A

letters or text or whatever uh brings a machine into that conversation without telling me. Right? So so on the point of transparency, if if I think I'm engaging with you as a human being, a fellow image bearer of God, and you start

Speaker A

copying AI output and I think I'm still talking to you, some fundamental deception has gone on there that I think is is quite awful and should be should be shamed. Um, but so I do think like you you have said, we do agree there are

Speaker A

legitimate and wonderful and productive uses. And so I think the transparency model makes the most sense. If you're going to submit a paper to a journal that involved whether on the research or writing side AI in some capacity, then

Speaker A

make sure that you are honest about where it came into play. And for professors who want to integrate it into their courses in whatever uh field, it should, I think, be the same way. you should have a some sort of explicit

Speaker A

permission of the instructor and some sort of open uh transparency about exactly what it did in the process as compared to to generated from your head.

Speaker A

So um I don't know if that fully answers the question but some appropriate uses of stigma and shame I I do support.

Speaker A

Yeah, that's good. Um you made a distinction between the printing press and the brother. We were asking you how do we know if AI is a good work or not? The printing press versus the brother. Um the one the with that particular one

Speaker A

obviously it's hard to think of a a pious use of a brother. I mean I'm open to any suggestions but I it's kind of hard maybe if it was empty and you were staying there the night as a as a a

Speaker A

invader or something but um but then the next one though was was you know penicellin versus fentinil. Uh and you know I guess the the question is how do you both of those have legitimate uses and illegitimate uses. So where where

Speaker A

would you I hate to this is a very difficult thing to answer. So, ju just your thoughts is fine. We're not trying to to win a debate here, but I'm just curious, how do you figure out um the

Speaker A

right way to use uh LLMs? Um and and where would you where would you what are the ones that you think are okay, but you wouldn't personally do, but if you thought if someone else did it, you would think it were morally listed like

Speaker A

using penicellin. Maybe maybe you're allergic to penicellin. That's not a good example. Maybe you are concerned about generating bacterial resistance in the world and you've made a commitment not to use penicellin in your particular herd of cattle of cattle or something

Speaker A

like that. So, you know, not that you think penicellin's evil, you just think it's unwise, right? So, I guess where is it evil? Where's using AI? Where is using an LLM um in your daily life evil?

Speaker A

And where is it just something that you wouldn't want to do because it's unwise?

Speaker A

Um the I'll do the second one first. The things I wouldn't want to do. Um I know people who use it effectively for and I have found it useful at times for um general background research in the sense

Speaker A

of pulling sources, research topics and questions. You know, give me the state of the literature on XYZ issue. And as long as you're using it in a way that actually provides you with links and you go and read the papers and things like

Speaker A

this, I think that can be quite useful. I generally don't do that because I prefer doing my research the oldfashioned way. It still kind of feels like cheating to just get the the AI to do my my annotated bibliography for me.

Speaker A

So um I I find that I I find more interesting things when I do my sort of exploratory research but at times I have resorted to it and I uh again as I try to avoid it. I know people who use it

Speaker A

quite well. That to me does not strike me as anything particularly bad. I just want to maintain those potentially atrophiable uh research habits that I already have of just I like just pulling things off the shelf and reading indexes and such.

Speaker A

Um evil uses. I think the the most obvious ones are anything that substitutes for real human community and relationship. So using it as a romantic partner, using as I would even maybe say a therapist, but you maybe we could

Speaker A

argue about that. Um but I I again back to the issue of it pretending to be human. I have almost been completely convinced that there is something disordered about having a conversation with these things uh because of the good of conversation

Speaker A

between human beings and and and what it should aim at. Um I'm not totally convinced of that, but I lean that way.

Speaker A

So that may be my strongest bit of pessimism. Um, and to that end, I would just say like all of those concerns that I have are like fentanyl, you know, great uses, but a real strong tendency to get you dependent. That same kind of

Speaker A

dependence seems to be uh resulting especially in consumer uses of of these models. I'm quite confident that people in in positions like you, the physicists I know, the biologists I know use it really well. I my students don't. So,

Speaker A

uh, can you specify what you mean by um conversation? So if you have a uh you come up with a research topic, you want to refine your ideas and you test them by putting putting a thesis into an LLM.

Speaker A

Is that the would that count as conversation? You're getting feedback from the feedback. I I would want to know what what sort of feedback uh is it's offering. Um like I can imagine, you know, just like I my spell check will

Speaker A

identify errors and that's a form of feedback. I would take that differently than the sort of substantive conversation I have with interlocutors at like an academic panel or even in a classroom. Well, here's an idea. Here's a text. What do you think about it? You

Speaker A

know, talk with me. Let's what ideas does this present? What feelings does it make you have? Those sorts of conversations I would not want to have with an LLM, if that makes sense. So, there are there are forms of feedback

Speaker A

that I can imagine being sort of uh technical and and specific and then others that I think are the more exploratory open-ended conversation that that we all value among our friends. And what I'm finding or seeing is that

Speaker A

there's a strong tendency for AI to give you a substitute for that that's not the real thing because it's not a human.

Speaker A

Could I could I move the discussion a little bit in a different direction? Uh you mentioned in your comments at the beginning, Philip, the uh effects of the automobile, unanticipated effects that remake the world once the automobile becomes

Speaker A

widespread. Um two questions and this is addressed to both of you. One one is uh in that kind of situation take the automobile what's what is the solution uh we have and look back in retrospect uh um you know smash the machine at the

Speaker A

beginning to prevent this from happening because they're there or recognize that the benefits are u you know what what's the solution then the second question is what uh where do you see uh AI potentially having the same kind of unintended consequences and

Speaker A

proliferating in ways that we don't currently anticipate? What what are the what are the worries you have about the ways it might remake our world uh in in negative in a negative fashion?

Speaker A

I'd love you to go first if you Sure. Um, no. Don't ask a man shaking his fist at clouds for solutions. That's way too practical. I I became a political theorist, so I didn't have to do real practical politics. Um, no. I uh

Speaker A

Russell Kirk's solution to his hatred of cars was to have his wife and his assistants drive him around everywhere.

Speaker A

He just refused to drive. So, you could just make somebody else uh bear the burden for you. Um, in in the case of something like Frankly, that doesn't seem like a solution. No, it's not. You still need you still need the roads.

Speaker A

Yes. Um, no. I I think you're rightly pointing out that there are certain things that are this like civilizational level that there is no real solution unless you make a choice to live in a situation that doesn't require you to

Speaker A

have a car. You can't solve the problem of other people having cars. You could move to a walkable community where you can use transit and cycle and things like this. You could make that choice if you wanted to. Um, for the most part,

Speaker A

it's uh asking people to sort of open their eyes to the harms attendant to the thing that might be inevitable as we've suggested. we might not be able to fix it in the sense of getting rid of it,

Speaker A

but maybe we can ameliorate the bad things that we didn't realize were being done in the first place. Um, as far as AI, yeah, I think it's it's going to have similar effects. It already has to the extent that um you you can't avoid

Speaker A

it no matter where you go. If you use any sort of technology at all, if you Google something, I was saying earlier I I I use this um e- in tablet. I like to scribble on it. It has a nice texture,

Speaker A

but if I want to send my notes to myself, I can send them in plain text and it uses a large language model to recognize my awful chicken scratch handwriting and turn it into plain text.

Speaker A

So even though I don't like using LLMs, I am every time I use this. Um, so I think it has already remade the world in many of those ways and what's left for us right now is to try to see what it's

Speaker A

doing as it's doing it and again those harms as best as we can. Um, yeah, that's uh I guess the Bible invites us to think about something like an idol and to think about how we become like the the

Speaker A

idol. That's one way you could go at this. So, what what aspects of idolatry um um come from looking at an idol like AI. Now, Psalm 115 talks about idols that have mouths but can't talk and you know, ears but can't hear and all that.

Speaker A

Um and if you serve them, you become like them, right? And so one of the things about idol worship is that people tend to that worship idols, they tend to go into idol service and you you go into

Speaker A

a building, you do your obesence to the idol and you don't talk to anybody and you go home and just go about your life.

Speaker A

It becomes this this transactional kind of thing. And so one one thing about AI, I do think it tends toward personlessness like you especially in the service economy, you know, you go into a place um the and and ensembles of

Speaker A

technology that's kind of one one thing in the technology literature. These these technologies all work together. So you take surveillance, you have a camera, you can interpret what's happening in in a room. The hotel I'm staying in right now, it has a snack

Speaker A

room uh that's that has no person in it. So it's it's completely empty of that.

Speaker A

But it does have a very sophisticated camera that can detect whether I'm taking a Snickers uh lawfully or unlawfully. And it even has a little thing where you can check yourself out, you know, and all that. So they have

Speaker A

this ensemble of things that enables personlessness in that environment. And and so think about any place in in life, especially in the service economy, where uh human judgment has been required in the past. Um I mean, we we started all

Speaker A

this a long time ago with a thermostat, you know, feels a little too hot, turn the AC on. Uh that's a very simple kind of judgment, but what about more sophisticated kind of judgments? When do I close a street down? When do I um when

Speaker A

do I sound an alarm? When is something dangerous happened or not? And we just take the people out of the situations more and more and more. and it'll be like it is in a lot of customer service situations where you just think if you

Speaker A

could just talk to a person um you could resolve the issue a little more easily.

Speaker A

Right? So I think that's one thing I see. It's the almost like uh pushing people out of the service economy in a way and and and like um like like the car it was Russell Kirk is that who was

Speaker A

so he has the privilege of wealth to to externalize his frustration here and be chauffeured around in some way. And that's the same way in the service economy. If you have the privilege you can access a a service economy that is

Speaker A

populated by people. And so I think you know a real tangible experience with real people is going to be at a premium and that will be very expensive and only a few people have it and most of us will

Speaker A

interact with intelligent systems all the time and never talk to a human. So that's one place where I can kind of see devil's advocate back.

Speaker A

Uh it wasn't too long ago when people complained about having nothing, no jobs except in McDonald's and now they don't need to take those jobs because they've been farmed out to AI. I mean there's there seems to be a you're taking up uh

Speaker A

assembly lines on in a factory. You got robotic uh robotics robots doing the work that used to be done by human beings. But it was deadening work. So u uh yeah how would you assess that that uh that

Speaker A

counter claim? I I think most of my like academic work is on the arguments being made about drudgery and mental mutilation in the factory system. People like Smith, Totoville and Mill talk about this and that is a very common response that you

Speaker A

get is well didn't technology actually alleviate this technological problem. the machines created the factories, but then we made better machines that freed those workers from that drudgery. Um, I think that may be a fair point as long as you have some understanding of what

Speaker A

those people are freed to or for. What what are they doing after they're freed from that menial job? Are they just going home and using a different algorithm while they, you know, doom scroll on TikTok or binge watch Netflix?

Speaker A

Okay. Well, they might also just be mentally mutilated by that, right? So, are you subbing, you know, one sort of uh physical harm for a kind of gross mental harm when you are kicked out of gainful employment? Um, but I think it's

Speaker A

it's it's true that we we've taken things that used to be awful for people and we've improved those uh industries significantly and that's we shouldn't denigrate that. Yeah.

Speaker A

Yeah. And and I think too the the um all the research right now is showing that it's the jobs that are the most automatable that are being you know lost but but the flip side of that is the

Speaker A

jobs that that can be more productive through augmentation. So like working alongside of a machine those are actually increasing in value. Um there's a a thing called a cobot where a human with great dexterity and and planning can work alongside of a robot that can

Speaker A

lift 600 pounds. And then you know as a pair of as an ensemble the human and the and the robot can work together in ways that neither could work alone. Um I guess the thing is how much of that

Speaker A

dexterity will eventually be on the robot side. And so this is the displacement of labor question to what extent so we we we uh you know other people in service jobs um and now there's there's not as many service

Speaker A

jobs. Where are they going? And I think this is that joblessness problem we're talking about. It it does point to me that we have to think about uh education from from first principles almost. I we we are in an education system that um it

Speaker A

seems incapable of um of producing the kind of output that we need. We we have especially in the United States um we push everyone to college um whether that's really the best thing for their goals or not. Um it can be

Speaker A

profitable for a college for someone to be there just for a few years. And so the fact that they won't be successful doesn't matter on the front end. And so that's really tough. We have kids uh getting a lot of just in case knowledge.

Speaker A

Like I remember in eth grade I learned all the different kind of uh sand dunes.

Speaker A

I know like a star sand dune and and that was a lot of just in case knowledge that I got in school. Um the the modern economy might call for people who are really good at getting just in time

Speaker A

knowledge and then applying it in immediate situations. So um I think we we're going to have to catch up. Right now, the big discussion in academia is cheating. But it's cheating within a system that wasn't working, that was

Speaker A

starting to peter out and it was dying. And so, how are we going to re how are we going to figure this out from the ground up so that people have a sense of calling earlier and go out into the

Speaker A

workforce and know what to do next? So, it's a deeper problem. Any further response to that on on the education front? I think you're right. Um the system was already broken and this just kind of revealed some of it. Um on

Speaker A

the how we do education it's very common to hear people suggest well what you should be doing now in your classroom is preparing students to use AI. I don't know what your experience has been but most of them at least in the general use

Speaker A

cases don't seem to have much problem using AI because it's quite easy to use.

Speaker A

I imagine in in your world your field there are specialized uses that they do have to learn in a classroom setting.

Speaker A

Um, but I'm just struggling to get them to to read a book and learning to use an AI is not going to help them read a book. So, yeah. Yeah. No, I'm with you there. I think that's a that's a big problem. Um,

Speaker A

um, it strikes me that right now we're we're talking about AI and getting kids to learn how to use it. We wouldn't have the same discussion about using everyday items. Like, are we going to make sure students know how to use mobile devices

Speaker A

so they can use their phones in the workplace? You know, that's not reasonable. And the internet, too. We used to worry about that with the internet. So, this this will pass too as well. I agree. Um although there are

Speaker A

some technical uses of it. Um but what about raising the standards? What about Okay, now that you have AI um what does it look like for someone who supposedly has already gone through you know 12 years of education and should possess if

Speaker A

if possible uh the discipline that was acquired in the absence of AI to read books and ideulate and think through things and now you get to college and now you're going to write a novel in a week or you know I'm I'm not saying we

Speaker A

should do that but I'm just saying what about just raising the bar. Um, but it would have to it would have to start by making sure students gain those skills earlier. And so that's why I think it's a a more architectural problem. But I

Speaker A

definitely resonate with you teaching them to use AI. It may be kind of a silly short-term uh thought that we all have. I don't know.

Speaker A

I'm going to call an audible here. Given the size of the group tonight, I'm not going to take a break and have you write down questions, but let's u go ahead and uh open it up to you all. Anybody has a

Speaker A

question you want to pose to both or either one of the speakers up here in Alabama. The legislature this past year passed the focus act which bans gadgets from K through 12. They all have to be hidden, nothing turned on. And um I

Speaker A

think that's a great thing that they did and maybe they will actually learn how to write a paper and um but I guess I'd like to see your opinion more on just the effect of all this time on children.

Speaker A

I think it's very Yeah. Yeah, I mean we went through a phase in K through2 education where the thing was giving kids Chromebooks and iPads and that seems to be fading because the measurable outcomes of that were quite bad. It doesn't seem like it

Speaker A

did anything to improve and if anything it harmed student learning outcomes if we're still looking at just outcome metrics. And so I think there's like obviously people like Jonathan height are talking about this a lot and have been at the forefront of pushing some of

Speaker A

these policies for phone free schools and things. Um I think we're gonna be seeing a lot more of that. right now it is kind of a boutique thing that uh is is more common at nicer schools and private schools but I think it's going

Speaker A

to become more and more common I hadn't heard about that but that's really interesting um yeah I mean I I have the blessing of being at a school where I think we get students that are doing better than some at other places that I'

Speaker A

that I've seen um but I I think especially post uh COVID and online schooling and things um the students who are technology dependent in the way that you're describing um they're just they're either not reading or in by some

Speaker A

studies not really capable of reading. Uh and if we could reverse that trend that would be great. It would make my life easier and it would make them better people. So yeah, I I don't disagree. I think I

Speaker A

agree. AI more gifts the art creative writing things like this. I mean earlier that begin it's hard to tell if it's real or not.

Speaker A

And so I just wonder if there's no restrictions on it. What kind of impact does that have on those particular fields in 50 years?

Speaker A

Um, are you an artist? I maybe you you speak to your music and then Oh, well, I don't know. Yeah, I really wish one of us was on the arts faculty.

Speaker A

That would be great if if one of us were an artist. But um I guess we're all artists in some way, so we'll go with that. Uh um yeah. Well, for one thing, I do think that uh the value of humanmade

Speaker A

things will may change. Um it may be that people are really excited to see um uh live performances again. that could be a possible side effect because you know that is a person standing up there sharing themselves with you. Um most

Speaker A

artists already are not making very much money from recorded music anyway and so they have to they have to depend upon live shows and that's where you you know get some intimate time with the person that you love in their in their art. So

Speaker A

that's important. Um I guess another thing just like schooling is a machine that has made AI's use inevitable. I think you know pop popular music is it's very formulaic. it's already written by, you know, seven or eight uh geniuses who

Speaker A

get sit around a table and write songs for an artist. And so that it already we already have a situation in which there's a machine-like quality to a lot of popular music at least. Um but you know, the other thing I would fear about

Speaker A

is demotivation. So an artist is just demotivated. I mean, look, if I can generate a beautiful photograph, um I've I've all my life I've had to make brochures that have pretty photos. And so I'd go to Adobe and I would buy a

Speaker A

piece of stock photography of engaging people sitting around a table talking about an issue to put in one of our ads.

Speaker A

And now when I go to the stock photography sites, they're all it's all AI generated photos because it's just it's too hard to to put a table together with all the permutations of human shapes and sizes and characteristics and

Speaker A

so you might as well just generate it by hand. But it looks it does not look like the real world in all the same ways. It may really change in the future. Um the other thing about creativity is that

Speaker A

very often an individual's voice is important. Like we want to see like you know the fact that um um sorry my brain is not is not chart but who just made the Odyssey um Christopher Nolan.

Speaker A

Christopher Nolan. People wanted to see what would Christopher Nolan do with the Odyssey. And so they didn't want to see just the Odyssey. There was something you know very specific about the artist that they would like to see and that

Speaker A

that probably will will continue. People will be inspired to emulate an artist. So I don't think it will kill artistic work. Um, you know, humans really uh start they love to embody the virtue required of of practice to become a

Speaker A

guitarist. The fact that there are are chess grand masters has never deterred anyone from becoming a chess player. Uh, the fact that that a million people on YouTube play guitar very well surely has not intimidated too many high school

Speaker A

boys from learning how to play guitar so they can sing a love song. And uh, you know, as long as love is your motivation, you're going to find a way to make art, I believe. So, uh, that's that's very important. So I'm I'm I'm

Speaker A

pretty sanguin. I guess the the the thing is if you if you really love pop music and you really want it to be made by humans, that might be one area where it's going to be very processed and and

Speaker A

already AI was used for autotuning voices and improving and polishing what was kind of human output. So just a few thoughts. Yeah, I think there's going to be as you described earlier a kind of emphasis or demand for authentic

Speaker A

boutique human generated content like you get your fair trade certified coffee you'll get certified humanowned or humanade whatever um I think that will exist and I think the real the worry on the motivation front I think is real um

Speaker A

because if you are wanting to do something as a profession or a career and it seems that those opportunities are drying up because those there are these easier options that could be discouraging um but you know it's similar to the analogy my language

Speaker A

faculty colleagues use with students is like you you could you know competently translate a foreign language text with these tools now very easily and read it in a fairly good translation you haven't learned the language and there's value

Speaker A

to learning the language yourself I think as you said if you if you love it if you enjoy playing music if you enjoy making art you're going to realize there's no real substitute for doing that yourself I also am kind of

Speaker A

speculating that there will be a resurgence of interest in more like tactile artistic crafts like woodworking is exploding right now and uh all sorts of places just things that aren't llmable I think people are going to start to value in that way

Speaker A

touching grass it's a it's a thing I'm um my question is about something you all didn't really touch on is the cost of AI on our planet you know the water all of that I've become I'm not a

Speaker A

big user of AI although I found like I Door Dash and things like that. Okay, I keep saying I'm not getting the human connection, but then I when I have the opportunity, I do not want to go to the

Speaker A

grocery store. So, I'm rambling a little bit. But then my thought is, but I'm giving that delivery person a job, right? Something to do. But in my daily, I use sometimes just for information and I get a a source.

Speaker A

So I've never used chat GPT or anything like that but lately I've become conscious of the the impact on our environment and our water and all of that. What do you all both of you um you know what do you all think about that

Speaker A

aspect of um I could start uh you know I think these kind of questions about resources really um one one way to think about them is is kind of globally you know the I guess think globally act locally that

Speaker A

that sort of idea the thing about is water resources are very local you know Memphis for instance where there's a controversial uh set of computers that Elon Musk put is one of the rainiest cities in the United States it has

Speaker A

gigantic river that runs right through it. So, you know, maybe maybe water wouldn't be an issue in a particular location, but maybe in Arizona it would be unwise to put a facility there. And I think the other thing, too, is that um

Speaker A

um most of us don't really think about how much water we use. It's really hard if someone tells you that this facility is going to use 20,000 gallons a day. It sounds, wow, that's awful. But most of us would be hardressed to say how many

Speaker A

thousands of gallons we use every day at our house or how how much water a Walmart Super Center uses or something like that. So, um, I think if you're trying to evaluate any of these things, my my advice would just be, uh, focus on

Speaker A

the local issue, you know, think think about this. Don't think globally on these things. You really need to think locally. Is water a problem here? Well, then maybe it's not the wisest place to put one of these. Um, electricity is

Speaker A

another one. Um, there's a lot of surplus electricity generated and it it is if it's not used, it's, you know, it's generated, the potential is there.

Speaker A

If you don't put it in a step down center and in a substation and then use it, it's just generated and not recaptured. we're not storing it in a battery. There are communities that have that are already producing surplus

Speaker A

energy which you could think of as a waste uh that could be used for something. So that's another thing. Uh in Canada there's lots of hydroelectric power. Um some countries have nuclear power with excess and all that. So maybe

Speaker A

electricity isn't a problem where you live either. Um and maybe if you had another um and because electricity is so highly regulated, it may not change your rates as a consumer at all in a particular area. So that's that's that

Speaker A

part. Um, the other thing is that these the when I'm looking at these news stories about data centers, I I'm careful to to listen to whether the problem is during construction phase or the using of the center because

Speaker A

construction is a terribly messy process. It's a violent process where you're drilling holes in the ground and pounding things and construction is pretty unpleasant to to be around. But then once the facility is written is built, you know, it's just a kind of a

Speaker A

run-of-the-mill big building that kind of hums a little bit, you know. So that that's another thing, the phase thing.

Speaker A

Um noise would be a bigger concern. Like if you uh uh and the noise comes in because if you're putting a data center somewhere that doesn't have sufficient electricity, a lot of them are putting behind the meter power generations.

Speaker A

They're putting generators in the parking lot and these are just loud. They beat their their gas turbines to generate power. And a lot of the controversy about the noise of some of the data centers in Mississippi is probably traceable to that. We just

Speaker A

don't have the electrical grid that's come up to speed in quiet ways. And so we've had to kind of temporarily use very noisy electricity. So um I that would be kind of I mean I'm we want to make sure that we use resources wisely.

Speaker A

Um water that is used for any purpose doesn't uh go into non-being. I mean it stays water it may become vapor and be recaptured later in the water cycle. Um it could be polluted in a way that makes

Speaker A

it more expensive to process. Uh it could be heated up in a way that makes it change the ambient temperature in a certain local area. All these issues have to be looked at and every data center is different. Um if we

Speaker A

overregulate you know that's another other thing we'll push them into areas where they may not be wise to put either um you know put them in space or or in the ocean. These are other um these are other kind of um theories and it's also

Speaker A

possible that AI just gets more efficient and it becomes less resource intensive uh to get the same amount of output or or what we need. So those are just some thoughts on that. I uh I'll just because I'm so negative about AI in

Speaker A

general, I have to work really hard to self-correct sometimes. So when people start criticizing these data centers, it really would feel great if I could just like confirm my priors and be like, "Yeah, look, there's another way that

Speaker A

it's awful." Um, the very little research that I've done, I think, is matches everything you've said. The one one thing I'll add is I've I've seen reports recently of cases where the data center construction ends up being a a

Speaker A

net environmental positive in places where um there have been like abandoned factories that the companies refuse to clean up the data center companies coming in and saying well we'll do the clean up and put the data center here or

Speaker A

taking over unused facilities like the Foxcon facility in Wisconsin. So there are maybe some positives out there as well um as well as the negatives. So yeah that's but the choice to build a data center is a choice about a particular future we're

Speaker A

going to have. it's going to be a where our investment is in AI. And so that's another if you don't believe AI is worth investing in, then that would be another reason to oppose data centers in general. But generally, you know, the

Speaker A

fact that, you know, I don't do Pilates wouldn't make me go down to the city council and oppose the building of a new Pilates uh building, you know, or or uh you know, if I don't personally smoke, you know, opposing a smoke shop or

Speaker A

something like that. I I don't know. So opposing businesses coming into a community because you don't personally like it, that's it's hard. So that's a morally aggressive thing to do. So hope that helps a little bit.

Speaker A

During the discussion, you mentioned new technologies around the way that we create society for them.

Speaker A

It's one thing to think of roads which aren't thinking and responding and changing. Uh what about the fact that we now have cameras all around us that are controlled by AI? I think of something like 1984 where big brother is always

Speaker A

watching you or like the panopticon. It seems like we're sort of creating our own panopticon. We have a phone that can be controlled by AI that can watch us.

Speaker A

We have block all around us. Every vehicle that comes out from here forward has to have camera in it that's always watching you via AI and the uh research shows that anybody that's under this sort of constant surveillance there is

Speaker A

drastic negative effects of that. So, like what are y'all's thoughts on this sort of is is this a real thing that I'm observing or is there is there maybe too much humorism inside of you?

Speaker A

Um, you know what's interesting about both the panopticon and 1984 angles is the panopticon is supposed to change the behavior of the prisoners, right? You could be observed at any moment. So, that changes how you act. 1984 as well,

Speaker A

it's at the beginning Winston says, you know, there's no real way to know if they're looking through the camera on my TV, but they always might be. So, I always act like they are. I always ask my students, you know, you guys have

Speaker A

seen these news stories about your phone listening to you, about cameras, about this and that, like, do you do you consciously change your behavior as a result of it? And they're like, nah, not really. Now, that may not be true.

Speaker A

You're right. Like, I've seen similar research what you're talking about with like anxiety and all these other things that develop as a result of this kind of surveillance apparatus. But I think one thing that's really interesting is just

Speaker A

like how passively we receive it. And maybe that's a bad thing. Um, I think we'll see like focusing incidents like have come out recently with police officers stalking people with flat cameras or people being misidentified on flot cameras that are really like the

Speaker A

data center issue going to focus some some public eyeire and that's I think perfectly reasonable, right? It's perfectly reasonable to show up at your city council meeting and say, "Why do we have this contract? What are you watching? How are you how's it being

Speaker A

used? Is it actually helping our community?" Those are fair questions. Um and and yeah, so yeah, I'm I stresses me out, but apparently it doesn't stress my students out. I don't Yeah, I I would agree. You know, any new

Speaker A

technology like take take policing, the car changes policing, too, because you go from the beat cop to the to the car.

Speaker A

Even if you have a beat that's on streets, you know, before before you have police and you had horse, you know, you had police on horseback before that that could go into certain areas and not other areas because they were on a

Speaker A

horse. Um, and then you have the ensemble of the police car and the camera that can dispatch you there.

Speaker A

There's also microphones that that detect gunshots, any kind of impulsive noise in public, triangulate where the sound is coming from, use AI to determine whether it's a gunshot or a backfiring car, and then dispatch a police officer to the site. And so I I

Speaker A

think all these things, you're right, the technology changes the way we do policing. And a lot of people depend upon a community's, you know, uh, sensitivity to that, if they really want to have that, if they want to feel

Speaker A

watched. Um, there's also been some kind of romantic backlashes, people spraying spray foam on these cameras and uh I remember there was I forget which political commentator said back in the early days of these drones that that someday there would be a folk hero who

Speaker A

shoots down a drone with a shotgun or something like that. And now that never happened. I don't think we have our John Henry or our uh or our lite who shot a drone with a shotgun. Uh we tend to

Speaker A

acquies to these things especially we think if we're law- abiding then what do I have to lose? But um you know that's that's the trap that we're always warned about giving up uh freedom for security or from or some other purpose. So I

Speaker A

guess it depends upon your value of freedom your uh expectation of privacy in a public place which is not typically what people expect and that's usually what the flot cameras are. So but I would say on the flot cameras that's a

Speaker A

private company as well. they have an interest in in you know being used by uh people not just municipalities but it's being used in places that don't have cars golf courses and other places and so you know they're going to be they're

Speaker A

going to be held accountable by the market as well I mean people may there be such a stigma around having them that they uh will naturally be less successful so I don't know there's other forces I guess of saying

Speaker A

one thing I want to add on that right so the another big previous you know surveillance state hub happened with the Edward Snowden stuff and the NSA and the metadata that they were collecting and one of the defenses of that surveillance

Speaker A

mechanism then was that it's um even when they had permission to look at your metadata it's really hard to glean a whole lot about you from your metadata that you know combing through it by hand is very time inensive and all these

Speaker A

things well these new technologies enable them to make much better use of that uh and turn that metadata into much more fine grain information about you so that's another source of anxiety perhaps same anxiety I think that obtains with

Speaker A

the flot cameras Well, I may not have an expectation of privacy in a public place. I might have an expectation that uh my local law enforcement agency doesn't intimately track my pattern through town, know where I stop and for

Speaker A

how long and who I'm visiting. That all feels a little bit weirder than just like, well, anybody could observe me instantaneously at any time. So, I mean, I'm sympathetic to to the concerns. I I teach a lot of sci-fi because I want

Speaker A

like you should be worked up about some of these things, you know, and then have good conversations about them.

Speaker A

And there's a conversation between the law and and these things. Like for instance with credit, you know, if someone denies you a credit application under the Fair Credit Reporting Act, you have to there's a right to find out what

Speaker A

information they're using to deny you credit. Uh the ability to challenge any kind of false information. So it's possible that law will catch up with some of these, you know, public uses of data. If they say you were speeding, you

Speaker A

know, there might be a higher burden to prove that it was you that was doing the speeding. And and so I I definitely could see that happening as well. There could be a legislative remedy that results in changing the law.

Speaker A

Any other questions? Yes, sir. You both talked about mechanical automation and AI taking over a lot created from automation require higher levels of technical expertise than jobs automation. So instead of assembling a bearing the maintenance machine bearing, how how much of the nihilism and

Speaker A

angueled by what seems to be a public or a tooling system that is failing training people for these higher higher tech jobs while we're moving to a world where those are like you have to have at least a two years of trade school to actually

Speaker A

get into the workforce. Yeah. Yeah. I think that's a a real part of it, right? this this fear that wherever you do get displaced from um there's not a ready job available for you. You can't do the new form of the

Speaker A

thing that's resulting. It's it's uh it's why I think you see figures like um Elon Musk and other tech uh billionaires suggesting trying trying to sort of prime people right now that we're going to need policy solutions to this. We're

Speaker A

going to need things like universal basic income to take care of these sorts of people who have these anxieties. Um, if you want a sort of like uh imaginative approach to this, uh, Kurt Vonagget's Play Your Piano was published

Speaker A

in 1952. And it's exactly this. The whole society is structured around if your IQ is a certain amount, you get sent to grad school, you get a PhD, the machines will listen to you. If your IQ is below a certain amount, you go, you

Speaker A

know, dig ditches and maybe serve in the army. But we provide for all your basic needs. You got an apartment, you got your food, you go live on that side of the river, and life is fine, except it's

Speaker A

not. It's boring and it's not human. So, it's a fun imaginative exploration of that. But I think the response to what you're describing is these people coming in and saying, "Well, we're going to find a policy solution to take care of

Speaker A

you when your job no longer exists because they're not optimistic that they can train you for those new jobs." There might have been some incentives, too. Maybe it was easier to get money for four years of education than to um

Speaker A

even just it was only just in the last year, I think, that you you could use a Pell Grant for a very short-term technical kind of education. So just even education funding favored um maybe more education than you needed for some

Speaker A

of those kind of jobs. So I don't know there there might be um and I think younger people are more open to um those kind of occupations. And so maybe um maybe it'll just be a higher esteem to

Speaker A

go out and and work with your hands or work with your mind and your hands in these very highskilled um occupations that are replacing the ones that are that are so automatable. Um but yeah, I mean that we have to we have to make

Speaker A

these viable options and show people how valuable they I mean, there there's some really good jobs like you're talking about. Definitely. I don't know. Do you do you I'm sorry I have to ask the audience member question, but do you

Speaker A

think young men your age are entertaining those options more than they used to? I think so.

Speaker A

I would think so. Um I'm also I'm in the Navy from the general. Oh yeah, no problem.

Speaker A

I'm going to close with a couple questions for myself and then I'll give you each a chance to make any final comments you want to make. You both agreed that technologies change us.

Speaker A

Um uh is that a problem? When is that a problem? How much of the change that happens with technological innovations did reskilling atrophy is a matter of division of labor uh where you have more complex uh economy wider distribution of

Speaker A

different activities and jobs. And therefore uh you don't have to have one person who's on the competence because is very competent in the one thing he can do and relies on other people to do the other thing. So um

Speaker A

when is it or is it a problem for technology to change us and how much of the uh how much of this atropy that we're worried about is just the division of labor things.

Speaker A

The the metric that I try to employ in assessing this is if the change comes at the expense of some good we consider to be higher. So if you know there are lots of ways I can imagine living differently

Speaker A

than I do. um and still being able to live the kind of life I want to be able to worship how I want to develop the skills that I want and so on. So like I could imagine living in a cabin without

Speaker A

air conditioning, but I have a home with air conditioning right now. I'm a different kind of person in each of those scenarios. Um I become very spoiled with my air conditioning, right?

Speaker A

So but that doesn't to me come at the expense of other things that I consider to be essential. Um I think you might find disagreement about what things are considered essential. That's fine. You can have multiple different ways of

Speaker A

approaching life. But there are certain lists, let's say, of virtues um that you ought to be cultivating in any scenario.

Speaker A

And if these changes wrought by technology are sacrificing those um then you might want to start thinking about it. So I don't know that's a conversation I would have probably within your your community, your church, your family, like what are the things we

Speaker A

think are the essential goods of human life that we ought to be pursuing and when has some piece of technology become enimical to that? And the model for doing this in a way that at least works for them is like the Amish. This is

Speaker A

literally what they do. They say it's not that we're opposed to all of it.

Speaker A

it's that these are the things we think are higher and so we're going to assess each new development in light of whether it undermines those things we think are better. Um, yeah, I don't if that fully answers the question.

Speaker A

Yeah, that that's the question is very perceptive and your answer is very perceptive. This is what LLMs are all about. We're going toward artificial general intelligence. That's the goal.

Speaker A

Completeness, um, encyclopedic knowledge of everything, omnimpetence, you know, but that and and you know, when I was a kid, my uh, mom wisely encouraged me to be well-rounded. go out for things, she would say, you know, join things. That

Speaker A

that was kind of her expression. And there's a there was a real wisdom to that. And um but that's really different. And I think that goes to she was trying to help me develop um wisdom and shrewdness and knowledge and

Speaker A

kindness and some of the moral and intellectual virtues that would be that would need to be present in any specialization. Even the specialization that makes you unfit to do the thing that you're now outsourcing or that you're looking as a division of labor.

Speaker A

And that's the thing about humanity. God has gifted us with a variety of interests. It's really hard to become good at something that you don't enjoy.

Speaker A

Um that you don't even have an inkling that you might enjoy it. Um and even music, you might have an inkling you might enjoy music, but it's one instrument versus another. And if we didn't have that, you wouldn't have an

Speaker A

orchestra with all the different kind of instruments. You wouldn't have a society if if God didn't help us to uh to fill all of these little ecological spots in the in the interests and in the kind of spectrum of people that we need. um

Speaker A

there's a diversity of people in a church, you know, that that that push it in a certain way that think of certain things that make the church's ministry more uh more more useful. So I think that's very perceptive that the division

Speaker A

of labor um we can't judge ourselves for outsourcing and technology might be an agent of doing that. And then you I think your response is great that it doesn't um it doesn't um acquit us from the the necessity to examine every

Speaker A

person and see if they're developing the virtues and whether the outsourcing has risked developing those. So well well-rounded is defined in terms of character instead of you know just knowing everything or or doing it all ourselves. I think I'm in my own life I

Speaker A

think I'm a little guilty of doing everything myself and not trusting others and I think that can be a lack of virtue to not give up things and and you know let my calendar remind me of things and you know use voice recognition or

Speaker A

something like that. My last is more of a comment than a question but I want to get your reaction to it. Um the the social uses of AI. Kids talking to their best friend is is an AI.

Speaker A

Um, it seems to me that's not a technical problem at base. Um, why doesn't this kid have somebody else, a person to talk to? Uh, are his parents still married? Does he have siblings?

Speaker A

Um, it seems like uh the technology gets blamed for something that's a familiar problem that's ultimately a spiritual problem and then the uh the technology is uh filling a void that's there for other reasons your reaction to that part which I guess

Speaker A

the other the other the flip side of this is if you think uh we need we need to make sure we fix the the or cut off the dangers of the technology, there's a there's a risk of becoming a technician

Speaker A

in trying to fix the technology rather than going to what could be a and I think in some cases definitely is a deeper spiritual root of the problem.

Speaker A

I think you're right that in some cases this is filling a void. Um may maybe in all cases depending on how we're defining that void, but I think there are cases where somebody with an otherwise healthy, happy family and

Speaker A

social life still finds these things. uh they they're drawn to them just the same way people might in you know a relatively loving marriage might be drawn into affairs and things like this depending on what else is going on in in

Speaker A

their life and one of the ways that I see this most is um in the context of scholarship and writing um where I very closely guard what I take to be uh the good thing that I'm doing from you know

Speaker A

having machines do it for me I talk to people who do use it and I ask them you well you have you have colleagues you have friends we can talk about your papers send me your paper I'll give you

Speaker A

comments why are you going to the AI to ask for them and their answers are things that are very similar to the things people with AI girlfriends say, "Well, it's always there. It gets back to me immediately. It never says no."

Speaker A

And so, like there's a very similar mindset in in both of these fields where I think it's the convenience, it's the ease, it's it is seductive in a way, even for people who otherwise I think are in healthier situations where I

Speaker A

wouldn't expect them to be drawn to that that it's it's the productivity, it's the output, it's the efficiency, and it's always there and it can always support me. It's very sickopantic. It will always tell me that that's the best

Speaker A

idea it's ever heard. That feels really nice. So, yeah, I think your question is, you know, it points to something true. Um, and I guess one of the remedies is to avoid uh the the temptation you mentioned is the

Speaker A

technical temptation. We'll just fiddle and make it a less seductive technology. Um, but maybe the the thought is just to make to to to stigmat well stigmatize, but just just to make it hard to prey on vulnerable people. you know, that's

Speaker A

sometimes we have laws that protect more vulnerable people from harms and maybe that's the place where um like the Bible um especially calls out um leading a person who can't see into a hole and there's a there's an added um moral

Speaker A

stigma moral I keep using the word stigma I apologize I sound like a chatbot you know just bringing that word out here all the time but uh there's a increased moral um pernitiousness about leading a blind person into a hole as

Speaker A

you're just for fun, joshing around or whatever. So, um, not only not only do you have a hole that was created by some other breakdown in a person's life, but then you have a person profiting from that breakdown in a certain way. And so,

Speaker A

that that's not a technical problem either. That's another and that's an active uh exploitation that we have to protect. And that's why I mean laws about data and children are different than laws about data and adults. That's why um you know taking away phones in a

Speaker A

school environment is way different than taking away your phone when you're at work. Um we we protect children. We protect um people who can't protect themselves. And so that's a that's a good principle for doing it. So that's

Speaker A

all I had to say about that one. Uh any final comments that you want to make? Um though I am I you know taking the pessimistic and negative tone I I do have great appreciation for people who are engaging with these tools in ways

Speaker A

that um are more more direct and active than I feel comfortable with doing myself. Um I think there are I know some Christians who are using them to great effect to uh great work in their callings and so I wouldn't want to

Speaker A

denigrate that. Um so I I just want to close with that. I appreciate the conversation. And I appreciate your your thoughts and your article and here and um appreciate Christians helping people who are using these tools think about

Speaker A

how to use them uh well for advancing the kingdom of God. So I really appreciate that. Thanks.

Speaker A

Thank you Phil. I really enjoyed the conversation. I was very productive. It was great to be with also talk to someone who deals with this every day as well. My my final thought would be about some of the uses of AI that that you

Speaker A

called out as being uh potentially problematic like asking for feedback on a paper and all that. And I guess the there's a virtue of humility that you know in the Bible. Um and one of the evidences of that is that we're called

Speaker A

to go to the ant to learn patience, you know. And so um I wonder if um if you've written something and uh and you think it's great, is it is it an act uh it is a is it a morally um negative act or is

Speaker A

it an act of the virtue of humility to submit it to an AI engine and say, "Hey, where am I weak here? Where could I where could I learn a little bit more here?" And uh and so I I think there

Speaker A

might be a place for that. I think uh and I think it it it is virtuous to be humble even uh to use a machine that um you know could be used improperly in a proper way in a humble attitude. So I

Speaker A

guess I'll leave it with that. Ask Brian Mo to come forward and we'll sing a final hymn.

Topics:artificial intelligencetheologytechnology impacthumanityethicsChristian perspectiveAI riskstechnology and characterfuture of humanityphilosophy of technology

Get More with the SozAI App

Transcribe recordings, audio files, and YouTube videos — with AI summaries and speaker detection. 30 minutes free.

App Store Google Play

Or transcribe another YouTube video here →

Free tools: TXT to SRT · SRT Validator · Merge SRT · Subtitle to Text · All tools