In this episode of Catholic Review Radio, Christopher Gunty explores with Taylor Black the rapidly evolving landscape of artificial intelligence, its cultural and ethical impacts, and the Catholic perspective on technology. Black is director of AI and Venture Ecosystems at Microsoft. Discover how AI is shaping society, the importance of human dignity, and what the future holds.

The conversation talks about the significance of Pope Leo XIII’s encyclical Magnifica Humanitas in the context of AI and technological shifts. Other key topics are the balance between technological advancement and environmental sustainability and the role of AI in developing nations and concerns around inequality.

Black also discusses the ethical considerations and the importance of maintaining human relationships in the age of AI, and makes some predictions for the next few years in AI development and its societal integration.

Host: Christopher Gunty

Guest: Taylor Black, Microsoft and Catholic University of America

News announcer: George Matysek Jr.

Producer: Christopher Gunty

Copyright © 2026, Catholic Review Media

Artificial intelligence is the new buzzword. Everyone is talking about it, and yet not everyone understands it. Even Pope Leo has weighed in on the topic in his first year as pope, releasing his first encyclical letter, Magnifica Humanitas, about the topic in May. So we're going to talk today with an expert on AI and the developments around it. Welcome to Catholic Review Radio. I'm Chris Gunty, Associate Publisher of Catholic Review Media. Our guest today, speaking from Seattle, is Taylor Black,

Director of AI and Venture Ecosystems at Microsoft. He's also the inaugural director of the Institute for AI and Emerging Technologies at the Catholic University of America. He was the keynote speaker. He was the keynote speaker at the recent Catholic Media Conference in Atlantic City, New Jersey. Welcome to the show, Taylor.

Christopher, thank you for having me.

Well w what do you do as director of AI and venture ecosystems? Are you thinking about AI all day, every day?

Yes, for better or worse. I I do two primary things in the CTO's office. One is I help early stage incubation efforts. So turning ideas into prototypes into b new business lines for Microsoft. And I also run our cross company peer reviewed conference motion. We have a a community of about 25,000 practitioners and we'll put on a peer- reviewed conference in various AI, ML and data science topics about once a month.

So you're the folks behind Copilot and other things like that?

Yeah, y yes, very much so. although there's there's a number of different teams working on and I kinda help knit all of their efforts together.

That's great. That's great. The pope's encyclical is titled Magnifica Humanitas. In English, that's magnificent humanity or the grandeur of humanity. Why for a document about technology is the emphasis on humans?

Well, because in the end this technology is a tool, just like many other kinds of tools that we've built. It can be a little bit hard to wrap one's head around it when it's chatting back to you and when, you know, the only other experience we've had of something chatting back to us is is a human on the other side of of that text box. And so I think it can be easy for us to pretend like it's human or think that it's more than a tool. but the pope is is here to remind us, like, no, we're the only things like

like ourselves and we need to lean into the richer aspects of understanding what it means to be human in order to really understand how to work with this technology well.

Yeah, and more and more these days if you ask Alexa or something, you know, you'll get back a very chatty kind of response rather than a very thing, I'm sorry, I d you didn't understand that. Let me try this. And so it's it's trying to be more human, but perhaps it helps to be reminded that it is not human.

Mm-hmm.

It does, yeah. I think it I think it makes our interactions better with the tools when we are working with them as they actually are, rather than something we we are pretending that they are.

Right, right. Can you give us a brief synopsis of some of the highlights of Magnifica Humanitas? What should people be looking for in that document?

Yeah, certainly. Well, the first part of the document I f I find of is a a beautiful summary of Catholic social teaching. Catholic social teaching kind of i is a is a wide and broad topic, but I think Humana Magnifica Humanitas does a good job of summarizing the major themes of Catholic social teaching at the at the beginning of the document. it then kind of goes into a bit of an exploration of of two major technological shifts as as

I'm I'm sure your listeners will will recall one of the reasons Pope Leo put it out when he did, and one of the reasons why Pope Leo chose his name, is to kind of mirror some of the activities that his predecessor Pope Leo XIII, who wrote Rerum Novarum and was wrestling with the technological shifts of the Industrial Revolution, and kind of taking that into into the current era. The two major things that he touches on as part of that technol that technological shift are

our AI, artificial intelligence as a as a technology that's kind of pervading every aspect of our life. and the other one is kind of this concept of transhumanism, which is this this technologically oriented belief that we through our own technology, through our own invention, can drop the bounds of mere mortals and become

like gods or live live for a very long time or or you know become something different than human. And he talks about the both of those things really thoroughly before kind of bringing us back into just the conclusion is just beautiful. I I wept the first time I read it. bringing us back into a reminder of of of how magnificent our our own humanity is as redeemed by Christ and as modelled of course so fantastically by the Theotokos, our Lady

and so yeah, that's that's kind of the trend line throughout the throughout the document and it's very much well worth a read.

Mm-hmm Why is it so important that Pope Leo XIV is speaking out and writing about AI now, right now, kind of in the midst of this development. As Leo XIII looked at the Industrial Revolution, it was well underway, in in some ways, and so he was speaking kind of behind the curve. It seems to me like Pope Leo XIV is out ahead of the curve on this, is he?

You know, that's a that's a good question. I I my thinking is he's about right where right where it needs to be in terms of timing. The thing that we have to remember is that this technological shift happens much faster than the industrial revolution could. you know, we already all we as the humans of the world already had cell phones in our hand, we already connected to the internet, and so when a model an AI model ships

All of us have access to it nearly instantly and and which is something very different from the technology of Pope Leo Xiii's time. and so I I do think that it it was a good timing. I think it's I think because this technology kind of hits us at a deeper place on our own kind of cognition and our own understanding of ourselves because of the way it touches our thinking. yeah, I I th I think it's I think it's timely.

Yeah. And certainly Pope Leo XIII could not have predicted when he wrote Rerum Novarum that that man would achieve space you know flight in 1903 and then land on the moon 60-some years later. So I mean that that development has been quick and yet the development in technology seems to be even faster as you as we go along. How is today's AI different from past search engines?

Mm-hmm, mm-hmm.

The early ones asked Jeeves, Bing, Google. I I mean, aren't they all just computers using you know, using the power to parse data? Or do the large language models that AI is based on really change the equation?

They do. They do actually. So AI has been around in at least as a term since the 1950s and it refers to a class of a class of really machine learning algorithms. And what machine learning is is really giving a mach giving a computer a set of instructions to compress and relate certain kinds of data within a training data set. there were three major technological shifts that made this AI

Mm-hmm.

moment really possible though. one is a paper came out I think in 2017 called Attention Is All You Need, and it's a i it was a a scientific paper that described a certain way of setting up an algorithm such that the algorithm could focus in on different parts of a data set in a in a deeper sort of fashion. The the second technological shift ha that happened was that

We NVIDIA went all in on GPUs, which is a graphical processing unit, which is the underlying hardware unit necessary for the training and development of AI. then the the third piece that happened is i is another kind of scientific paper that allowed a a bit of a recursion in the algorithm to happen, which is being able to move things that the algorithm had learned in earlier stages of the training.

through to later stages of the training so that it would get better and better over time. What those three technological shifts allowed is the ChatGPT moment of you know, tw November twenty, twenty two, tw to one? I I'm forgetting the the year, of course. but what happened there is that we are able to take basically everything that humanity has ever written, in the top twenty languages that we've ever wr written it in, compressed it and then

Mm-hmm.

found all of the statistical relationships with regard to any piece of data within that massive corpus. What that means now, instead of like where a search engine kind of w does a something called indexing, it it it notes where all of the information in a particular in the web is, for example, what AI does is it it not only kind of t explains where to find a certain piece of information, but it is able to

do that in relationship to every other piece of information within its data set. And so what that enables is things like, you know, I could I could say I stubbed my toe. And the AI would know that there was a a feeling of pain related to that. It was maybe a feeling of frustration related to that, because every time that anybody has talked about stubbing their toe within the corpus of that information, that that feeling of pain or that feeling of you know frustration came along with it. And so

As a result of that statistical relationship that happens in something tech calls hyperdimensional space, we're able to understand the relationships between data within the corpus at a much much higher rate. and the size of the amount of information that goes into it also ended up being kind of being a a unlock in terms of the training of models.

As we added more data to the model, the more capable and more intelligent they became, and leading to the the incredible effects that we're we're able to see today.

Mm-hmm.

Yeah. I in nineteen eighty one when I was a college senior I wrote a short story that got published in the literary magazine about essentially about robots in a factory deciding to unionize. b because they they were experiencing all these things and then they started asking for days off and things like that. it it seems like that was probably a precursor to to some of the kinds of things that we're looking at today of of just we we wonder how smart these machines are and

cool.

and and what they know and what they what they don't know. So yeah, that that's exciting. well we're we're gonna take a quick break and after the break we're gonna talk some more with Taylor Black, who is director of AI and venture ecosystems at Microsoft, about artificial intelligence and how we relate with it in the world. This is Chris Gunty, you're listening to Catholic Review Radio.

Yeah.

Welcome back to Catholic Review Radio. Our guest today is Taylor Black, Director of AI and Venture Ecosystems at Microsoft and inaugural director of the Institute for AI and Emerging Technologies at the Catholic University of America in Washington, D.C. Taylor, we we were talking about the Pope's encyclical Magnifica Humanitas. In the introduction, he writes: "New technologies open up a horizon extending in directions which are imaginable but not yet fully predictable.

This complicates the assessment of their potential impact and the long-term effects they may have on both the dignity dignity of individuals and the common good." What's the difference between imagining the future of AI versus predicting the outcome?

Yeah, certainly. Well, I I think a good analog is looking at cell phones, for example. you know, when when cell phones first came out it was it was kinda like having a superpowered walkie talkie, you know, where y you could talk to another person and walk around in your backyard without having to be tethered to, you know, the the phone. But of course we didn't realize that also

Right.

Bringing social media into our lives through the cell phone and kind of having it pervade the fabric of our family or the the fabric of of our own thinking when we're, you know, awake at three AM in the morning and the cell phone is there kind of as a re easy conduit to a to a world that we hadn't encountered thus far. and and things like even like Uber, for example, where you know

as a result of the cell phone, we're able to call up a stranger to pick us up in a car and and and and take us there. Both of those things w didn't weren't weren't obvious from the invention of the cell phone. And so I think we have a similar sort of thing here with regard to artificial intelligence where th while we have a lot of possibility for what artificial intelligence can do, where it can take us, both from a you know, workforce sort of standpoint, from a

intellectual companion and research companion sort of standpoint, to kind of revolutionizing how we even think about software in the first place. All of those are great possibilities that we still have to drive down to actually implementation. not only implementation in the sense of we got to ha we have to solve the engineering problems necessary for this technology technology to work, but also from a

How is this impacting us? Because every tool that we use ends up shaping us in some fashion too. And so what what are the ways in which this technology is already doing that and how do we understand that and shape that better?

Yeah.

Yeah. And your your Uber analogy, you know, kinda goes into Waymo and other things like that where you've taken the cell phone and the Uber and combined that with artificial intelligence to now drive me around town. So and and we rode in a f in Waymo's a few times last year in Phoenix and

I think the experience was pretty good and and I was I trusted it, which was an interesting thing about it. I heard somewhere a while back that it you know, somebody saying, I I don't want AI to do my writing so I have more time for laundry. I want AI to do my laundry so I have more time to be creative. do you think humans will get lazy intellectually and physically if they rely too much on AI?

Well, you know, we we have already with regard to television and with regard to, you know, driving everywhere instead of walking. So I think I think it is a a danger that comes with any technology that we use. Now but your point to why AI in particular might be a little bit more insidious is that it can do some of our thinking for us, right? So if we can hand off parts of our thinking, which is kind of a deep hits us at a deeper level, you know.

While walking is important for humans, it's it's perhaps a less uniquely human attribute, right? My my dog walks, my cat walks. but but the thinking piece is something that is more specifically human, and and and the pope calls that out in the encyclical too, of of not getting rid of that intellectual labor in service of efficiency. He even goes one step further, I think, and says, taking care to not

give up relationships with other people as a result of being able to use this technology. Not that not that they will completely replace it, but it will reduce the demand one has for seeking out companionship and seeking out relationship because we're able to answer certain kinds of questions with the tool.

Right.

Sure.

Yeah. There is a certain and some people say significant environmental impact from the AI data centers. You said at the Catholic Media Conference that AI will always be getting more efficient than it is right now. But even so, as the use of AI increases, how do we balance these competing demands of energy, land use, technological advancement?

Yeah. Yeah, I mean it's a complex question. I think it ends up being almost site specific, right? being able to put up a data center in in you know, the Mojave Desert gives you access to a bunch of clean energy, but then you have water concerns and and kind of vice versa. And so I think it is really is a a complex question that that requires a lot of thought not only from the hyperscalers and from the regulators

But also from the communities that are part of these decisions and and they should be brought into these decisions. Microsoft in particular has a really great, I think, forward thinking policy around how they are going about data center development themselves. and and I'm excited to see further efforts in that area. But but I think the underlying thing too is is understanding the value that AI is realizing.

So we can think of it as this why am I, you know, why are we setting up an environmentally problematic data center for me to generate fun images on my phone, right? and and that's not a great use. but other great uses though, like you can also use the same technology to better detect cancer in in biopsies and cancer screens. You're able to come through with

Right.

scientific discovery and and breakthrough as a result of utilizing the these data centers in the in in a fa in i in those sorts of ways too. And so it's it's a balance. It's a balance. And I think I think it's part of our own decision of what we're using AI for, but also being part of the community that's making those

Yeah, and you've you've mentioned at the Catholic Media Conference that you know we've we've come to accept other environmental trade-offs. Jet travel itself has a huge environmental impact and yet we have accepted it as a way to get more easily, more quickly from point A to point B. So there is that that kind of thing going on. we also know that poorer communities often feel the impact of new technology more f more profoundly. They often lack access to the technology itself, but they experience more of the environmental downsides.

How can we help mitigate those discrepancies?

Yeah, well there's also some silver linings to this as well though, I will say. we're seeing a huge uptick in developing nations with regard to entrepreneurial activity because all of a sudden people are able to get answers to specific kinds of how do I form a company or how do I think through this legal problem or how do I think through this HR problem? and they have a companion that's very articulate in their own language where they're able to get those kinds of answers more quickly.

And and as a result have more economic activity. But to your point though too, from a subsidiarity standpoint, and Pope Leo talks about this in the encyclical as well, of ensuring that the communities that are impacted by decisions also have a voice in the decision making process. I think there's still a tremendous amount we can do in in those veins of ensuring that you know the marginalized have

Mm-hmm.

have some say and have some see some benefit from these technological advances as well.

Yeah, that's critically important, I think. You know, just in the interest of full disclosure, the recording platform we use for this radio and podcast uses artificial intelligence to edit audio and video clips, clean up background noise, stuff like that. But for those people who want to reduce their reliance on AI, what are some ways to do that?

Yeah. this is a great question. I I like introducing people to the concept of and it's a it's a it's a $50 word, but it just means thinking about your thinking. But your your metacognitive strategies of how you come to the technology, right? So you can imagine you can imagine, you know, sitting sitting in your chair in the morning and opening up chat GPT and just throwing in a couple of random questions in where you who really haven't examined what you're using that tool for.

Uh-huh.

Maybe it's actually distracting you from something more important, like your children, or maybe prayer or something. And so rather than approaching it in kind of this this accidental or or you know, cursory sort of way, another way of approaching this would be like, wa well, this tool this is expensive from an environmental standpoint, perhaps, or at least I want to treat it as such. It's also expensive from a me giving it attention and and it having to work with it in my own intelligence, so

Right.

I'm gonna use this tool very specifically for something like when's you know when's the best time of year to visit Phoenix or like to do a thoughtful research project or a thoughtful set of activities that will help me get to better answers about something that I'm trying to solve in my own life. Now, thinking through those things, th this is the metacognitive part of it, where you're thinking about how you're going to approach the tool and the kinds of outputs that you want from the tool and

kind of being circumspect in how you're engaging with it yourself, that not only ensures that you're approaching it from mo a more sober point of view or a more thoughtful point of view with regard to your own intelligence, but it also ensures that you're using the tool for something substantive value wise for yourself, where you're able to kind of understand that yes, I am you know, sp causing these environmental impacts, but it's for this set of activities that I want to get done.

And it's useful in that context.

So kinda like an AI budgeting.

Exactly. That's a great way of phrasing it. Yeah.

I I like that idea. So what are you teaching your students in your courses at Catholic U in Washington? Are are these technological courses or do they also focus on the theological and teleological aspects of artificial intelligence?

Very much all of the above. in fact. particularly I I focus I'm I'm faculty in the business school and in the computer science department. and right now I'm focusing kind of in two major areas. One is is this sort of pedagogical concern. How do we think about AI as a as a learning companion or should it be a learning companion? And how do we how do we wrestle with the the more minute aspects of that as we introduce this technology into the classroom? the other

Yeah, there were days when we couldn't bring calculators into our into our classrooms, you know, when we were taking a test, so

Yeah. Exactly. Well and I I think it's we're we're approaching it from a similar sort of vein, here, you know. because it can do your thinking for you and in college is a time to learn how to do your thinking, y you do have to be careful with that. and and not throw the throw the proverbial baby out with the bathwater. But the other thing that I I I teach there is technology commercialization.

Yes.

so how do we take all of the innovation and advancements that that this technology and others can provide and how do we how do we turn them into companies? How do we turn them into license licenses or other kinds of revenue streams for the university?

Okay.

Great. Well quickly before we wrap up, what's next in the coming years of AI? Or are we even looking at it in terms of years? Are we doing month to month or day to day?

Yeah, well I'm excited maybe for absolutely for for three things. One is I'm I think that we've this technological development time has been a lot of fun, but I'm really excited to see the landing of the of the value proposition of artificial intelligence and all of the varieties of areas in which we can do that. I'm I'm also excited to see the tech the scientific advances that we'll be able to make as a result of having

Mm-hmm.

This companion intellect that can help us solve hard problems in in really powerful ways. and I'm also excited the way for the way in which AI is driving infrastructure development in the United States in particular, not only from a power grid standpoint, not only from a heavy industry standpoint, but also from from a kind of deeper understanding of the interconnectedness of a lot of this in order to produce kind of

Right.

grid level effects of for for United States industrialization.

That's awesome. Well, this has been a fascinating conversation about artificial intelligence and the pope's encyclical Magnifica Humanitas. We have been talking with Taylor Black, Director of AI and Venture Ecosystems at Microsoft and inaugural director of the Institute for AI and Emerging Technologies at the Catholic University of America in Washington DC. Thank you so much for being with us today, Taylor.

Great to be here, Christopher. Thanks for having me.

Thank you. This is Chris Gunty of the Catholic Review, and you've been listening to Catholic Review Radio. Don't forget to like us and subscribe on your favorite podcast app.