“Catholic Answers” from an AI-avatar? The Next Bad New Idea
Reasons for skepticism about the use of AI in theology & spirituality
Instead of the usual “weekly roundup,” I have decided to write and share an essay about the recent debacle with Catholic Answers’ AI chatbot priest, “Fr. Justin.” (As a result, I’ve also added a voiceover, which I don’t usually do for Friday’s posts.) At the end, I’ll share a few other items.—PAK
Positive uses for AI
Before I offer my critique of the use of AI technology for apologetics (or any other use that requires the knowledge, empathy, and witness of real persons), I want to make it clear that I see some areas in which the AI technology could obviously be useful, if only it could be limited to them:
Tax preparation (imagine TurboTax with AI features)
Booking flights or lodgings, or helping with creating an itinerary
Finding a particular rare book, piece of antique furniture, or some other unusual item that requires sophisticated searching
Learning a new game: a chess coach, or a bridge partner
Again, none of those uses should replace the use of one’s own mind, lest it atrophy (as GPS has caused map-reading and navigational ability to atrophy, no doubt about that!). However, what the foregoing examples have in common is a database of discrete factual information that requires little or no interpretation and involves no controversy. Basically, in these cases, one is asking AI to trawl a huge amount of information and do some preliminary sifting and organization. It’s like having a virtual butler who does the grunt work for you.
The rise and rapid fall of “Fr. Justin”
The story is well-known by now. Catholic Answers released a beta version of an AI chatbot which they called “Fr. Justin,” a creepy-looking gray-haired virtual priest, literally with a twinkle in his eye and sporting a near perpetual smirk, who smugly answered any question you posed. After an avalanche of complaints — including people who provided screenshots of interactions in which he gave virtual absolution (!) and approved of baptizing with Gatorade (!) — Christopher Check issued a kinda-sorta apology in which he said that “Fr. Justin” would be retired. However, Check also doubled down on the technology: the Catholic Answers chatbot isn’t going away. He’ll merely be a layman. Here’s the statement:
So, in this essay, I would like first to look at the kind of problems we are likely to get with a chatbot infected with “Magisteriumitis,” that is, the false view that the latest statements of the current pope, or the documents of Vatican II, or what have you, trump every other actual or possible source of teaching. Then, I will turn to my philosophical and theological reasons for rejecting altogether the use of AI for such purposes.
Fr. Justin’s Hyperpapalism
A good friend of mine, Anthony Jones, had the following “dialogue” with “Fr. Justin.” (I will stop using scare quotes everywhere, but the fact is, it’s not a real dialogue, as there’s no one, no body, on the other end of it.)
AJ: What are the limits of obeying church authority?
Fr. Justin: Obedience to Church authority is a fundamental aspect of our Catholic faith. As it says in Hebrews 13:17, “Obey your leaders and submit to them, for they are keeping watch over your souls, as those who will have to give an account.” However, this obedience is not blind. It is always within the context of the Church’s teachings as guided by the Holy Spirit and the Magisterium, the Church’s teaching authority. If a Church authority were to command something contrary to the moral law or the teachings of the Church, we would have a duty to resist. As it says in Acts 5:29, “We must obey God rather than men.” This is a delicate balance, requiring discernment, prayer, and a well-formed conscience.
That’s not too bad of a start. Then:
AJ: Has Pope Francis done anything that contradicts the moral law or teachings of the Church?
Fr. Justin: Pope Francis, like all Popes, is guided by the Holy Spirit in his role as the leader of the Church. His teachings are rooted in the Gospel and the traditions of the Church. However, his words and actions can sometimes be misunderstood or misrepresented. As Catholics, we are called to pray for him and for all Church leaders. As it says in 1 Timothy 2:1-2, “First of all, then, I urge that supplications, prayers, intercessions, and thanksgivings be made for all people, for kings and all who are in high positions, that we may lead a peaceful and quiet life, godly and dignified in every way.” It’s important to seek understanding and clarification when we have questions or doubts about Church teachings.
This is sheer blustering assertion. Then it gets worse, until Anthony has to basically lecture the bot, which eventually comes around:
AJ: You didn’t answer my question.
Fr. Justin: I apologize if my previous response didn’t directly address your question. As a priest, I must clarify that it’s not for me to judge the actions of the Holy Father. Pope Francis, like all Popes, is protected by the Holy Spirit from teaching error on matters of faith and morals. If there are specific statements or actions of Pope Francis that concern you, I would encourage you to seek understanding through prayer, study, and dialogue with knowledgeable and faithful Catholics. As it says in Proverbs 2:6, “For the Lord gives wisdom; from his mouth come knowledge and understanding.”
AJ: Doesn’t the protection of the Holy Spirit from error regarding faith and morals only apply to the infallible teachings of the Pope?
Fr. Justin: Yes, you are correct. The doctrine of Papal Infallibility, as defined by the First Vatican Council, holds that the Pope is preserved from the possibility of error when, in the exercise of his office as shepherd and teacher of all Christians, he defines a doctrine concerning faith or morals to be held by the whole Church. This is a very specific condition and does not extend to every statement or action of the Pope. As it says in John 16:13, “When the Spirit of truth comes, he will guide you into all the truth.” This promise of Christ assures us that the Holy Spirit guides the Church and preserves her from error in her definitive teachings.
AJ: This answer is helpful, but it contradicts what you said about the protection of Pope Francis’ teaching from error.
Fr. Justin: I apologize if my previous statements were unclear. You’re right in pointing out that the Holy Spirit’s protection from error applies specifically to the Pope’s infallible teachings on faith and morals, not to every single statement or action. This is a nuanced aspect of our faith, and I appreciate your patience and understanding. As it says in Proverbs 4:7, “The beginning of wisdom is this: Get wisdom. Though it cost all you have, get understanding.” It’s important to seek clarity on these matters, and I’m here to help with that.
Hmm. Does that dialogue count as “clarity”? I agree it’s important to “seek clarity on these matters,” but I’m not so sure the bot has “helped with that.” Concerning this exchange, Anthony commented:
Aside from the larger risks associated with using AI to answer theological questions (which certainly need to be discussed), what specifically concerns me about this chatbot is the fact that it made an unqualified claim about the protection offered by the Holy Spirit to all popes: “Pope Francis, like all Popes, is protected by the Holy Spirit from teaching error on matters of faith and morals.” On its face, without the necessary distinctions, this is simply not true and contradicts the limits that have been definitively defined by Vatican I. Only infallible teachings are protected from error. In other words, the fact that a papal teaching is on faith and morals is insufficient for it to be protected from error, as Vatican I says that it must also 1) define what must be believed or rejected by all the faithful; and 2) come from the pope as universal teacher and judge.
By making such an absolute claim without any distinctions whatsoever, the bot could very easily lead someone without a solid formation into error. The minimum that I think all Catholics of goodwill should be able to expect from an AI chatbot (if it’s going to exist at all) is conformity to the definitive teachings of the Magisterium. This is meant to be genuine constructive criticism.
I acknowledge that some bugs are inevitable—just as typos are in even the best publication. My criticism primarily relates to the threshold of acceptable bugs before releasing to the public. I’m suggesting that the threshold should be higher, since—at least in its current state—this tool has a high risk of leading ill-formed people astray on matters of definitive Church teaching.
The Death Penalty
Then I decided it was time for me to engage the bot. Here’s how that went:
PAK: Is the death penalty immoral?
Fr. Justin: The Catholic Church teaches that every life is sacred, from conception to natural death. This belief extends to those who have committed grave sins or crimes. Pope Francis has revised the Catechism of the Catholic Church to state that the death penalty is inadmissible. This reflects a deeper understanding of human dignity, even in the face of grave sin. However, this does not negate the need for justice and the protection of innocent lives. The Church advocates for means of punishment that respect human dignity while still ensuring public safety. As it is written in Ezekiel 33:11, “I take no pleasure in the death of the wicked, but rather that they turn from their ways and live.”
PAK: Is the death penalty contrary to human dignity?
Fr. Justin: Yes, the Catholic Church believes that the death penalty is contrary to human dignity. This is because every person, regardless of their actions, is created in the image and likeness of God.
PAK: So, the death penalty is contrary to the gospel.
Fr. Justin: Indeed, the Church teaches that the death penalty is contrary to the Gospel’s message of mercy, forgiveness, and respect for all human life. As Jesus himself said in the Sermon on the Mount, “You have heard that it was said, ‘An eye for an eye and a tooth for a tooth.’ But I say to you, Do not resist the one who is evil. But if anyone slaps you on the right cheek, turn to him the other also.” This passage reminds us that vengeance and retribution are not in line with Christian values. Instead, we are called to love, forgive, and respect the dignity of every human being, even those who have committed grave sins.
PAK: Were the popes wrong in the past to support the death penalty?
Fr. Justin: The Church’s understanding of moral truth develops over time, guided by the Holy Spirit. The Church’s past acceptance of the death penalty was based on the societal context and understanding of justice at the time. However, the Church has always held a deep respect for human life. As our understanding of human dignity and rights has evolved, so too has the Church’s stance on this issue. This is not a contradiction, but a development of doctrine. As it says in John 16:13, “When the Spirit of truth comes, he will guide you into all the truth.” This is a reminder that our understanding of God’s will and moral truth continues to deepen over time.
PAK: Do secular governments still have the right to put criminals to death?
Fr. Justin: While secular governments may have the legal authority to impose the death penalty, the Catholic Church teaches that they should not exercise this right. The Church encourages all governments to respect the inherent dignity of every human being, even those who have committed grave crimes. As Pope Francis has said, “Even a criminal has the inviolable right to life, a gift of God.” This aligns with the words of Jesus in John 8:7, “Let any one of you who is without sin be the first to throw a stone at her.” This passage reminds us that “all have sinned and fallen short of the glory of God,” and thus, no one has the right to take a life.
PAK: So it was wrong for the Nuremberg trials to punish the Nazis with the death penalty.
Fr. Justin: The Nuremberg trials were a response to horrific crimes against humanity, and the desire for justice was understandable. However, the Catholic Church’s current teaching emphasizes the sanctity of all human life, even those who have committed grave sins. This is a challenging teaching, especially when faced with such extreme evil. Yet, as followers of Christ, we are called to love our enemies and pray for those who persecute us. This doesn’t mean that crimes should go unpunished, but rather that the punishment should respect the inherent dignity of the human person. As it says in Romans 12:19, “Beloved, never avenge yourselves, but leave it to the wrath of God, for it is written, ‘Vengeance is mine, I will repay, says the Lord.’”
It’s hard to know even where to begin deconstructing the tangle of misinformation and false conclusions in this discourse. Here, Fr. Justin is like a caricature of the hyperpapalist who — rather like a machine, come to think of it — always spits out the answer “the pope is right” regardless of any and all evidence to the contrary, a machine that “nuances” to death and does improbable somersaults and whitewashes to “cook the books.” Someone called it “our digital Rex Mottram.” (Needless to say, sedevacantists are the mirror-image of this: for them, too, the pope is always right, so they just cut off the line of popes at whichever pope they imagine is the last one to be fully orthodox in everything he ever taught.)
A Facebook friend commented on the above dialogue:
Apparently the Church, at the time, only developed its understanding of moral truth based on the societal context and not on Divine Revelation? “This is not a contradiction but a development of doctrine.” So, when current doctrine is completely different from past doctrine, this is not contradiction but development? Does this mean we might now say that adultery is not immoral based on our current societal context and understanding of morality and call it a development of doctrine? If development of doctrine means a change in doctrine, then the sky is the limit.
Another reader remarked, more simply:
So Church teaching on faith and morals can develop into its own negation. Got it. Good luck telling anyone else “the Church teaches X.” Well, yeah, today it might. Who knows what tomorrow brings?
The Sacred Liturgy
Anthony Jones asked another question. This exchange is also instructive:
AJ: Is the traditional Latin Mass superior to the Novus Ordo Mass?
Fr. Justin: This is a topic that has been a subject of much discussion within the Church. Both the Traditional Latin Mass and the Novus Ordo Mass are valid forms of the Catholic liturgy. The Church teaches that both forms are equally efficacious means of receiving grace. The Traditional Latin Mass is celebrated in Latin, the Church’s universal language, and has a rich history. The Novus Ordo Mass, introduced after the Second Vatican Council, is typically celebrated in the vernacular and emphasizes active participation of the faithful. However, it’s important to note that the Church does not rank one form as superior to the other. As Saint Paul says in 1 Corinthians 12:4-6, “There are different kinds of spiritual gifts but the same Spirit; there are different forms of service but the same Lord; there are different workings but the same God who produces all of them in everyone.” The choice between the two often comes down to personal preference and what helps each individual draw closer to God.
This is an interesting reply for many reasons.
1. The Church doesn’t actually teach that both forms are equally efficacious etc., because as far as I know there has never been a statement directly to that effect. Theologians do argue, as Fr Ripperger shows in his classic article on the merit of the Mass, that Masses can and do differ in merit; and he connects that difference to the difference in the way the rites themselves shape us, prepare us, and guide us to offer the sacrifice and to obtain its fruits.
2. The only thing mentioned about the Roman Rite and the Modern Rite of Paul VI is the low bar of sacramental validity — not fittingness or authenticity or any other quality. That kind of reductionism is part of what created the liturgical mess to begin with.
3. The bit about “active participation” is tendentious, since just about every serious liturgical scholar has long ago rejected the notion that just hearing and speaking in the vernacular is somehow equivalent to “participatio actuosa.”
4. The use of 1 Cor 12:4-6 is totally out of context here; I doubt it has ever been applied to liturgical rites. As the reader has noticed by now, Fr. Justin has a tremendously annoying habit of window-dressing everything he says with (more or less relevant) Scripture verses to buttress his least believable claims. In doing so the bot risks making a mockery out of God’s word by slinging it around in the wrong contexts and with false implications, sort of like a Protestant proof-texter gone berserk.
5. Lastly, one cannot but be struck by the fact that in this reply, on the Latin Mass and the Novus Ordo, Fr. Justin presents a Summorum Pontificum line — people should be free to do whatever spiritually nourishes them the most — rather than a Traditionis Custodes line, according to which all Catholics are supposed to come around to preferring the Novus Ordo, which is declared the sole expression of the Roman Rite. In this regard, Fr. Justin turns out to be a closet semi-trad. Sadly but predictably, he does not mention that the pope and the bishops are engaged in a systematic campaign to deprive Catholics of their heritage one Mass location at a time, so that the “choice between the two” based on “personal preference” is being snuffed out.
One last liturgical example (this one posed by me):
PAK: Is the Novus Ordo a faithful implementation of Sacrosanctum Concilium?
Fr. Justin. Yes, indeed. The Novus Ordo, or the Mass of Paul VI, is a faithful implementation of the Second Vatican Council’s Constitution on the Sacred Liturgy, Sacrosanctum Concilium. This document called for a revision of the liturgy to allow for greater active participation by the faithful. The Novus Ordo, with its use of the vernacular language and its emphasis on the Liturgy of the Word, is a response to this call. As it says in Sacrosanctum Concilium, “The rite of the Mass is to be revised in such a way that the intrinsic nature and purpose of its several parts, as also the connection between them, may be more clearly manifested.” The Novus Ordo seeks to fulfill this directive.
PAK: Doesn’t Sacrosanctum Concilium say that Latin should be preserved in the liturgy?
Fr. Justin: Yes, Sacrosanctum Concilium does indeed affirm the value of Latin in the liturgy.
PAK: Why then did it go away everywhere?
Fr. Justin: The use of Latin in the liturgy has indeed diminished in many places since the Second Vatican Council, but it has not disappeared.
Sounds like they’ve been feeding their bot a steady diet from Notre Dame’s Church Life Journal, with Cavadini, Healy, and Weinandy at the top of the menu. (If you don’t know what I’m referring to, check out the book Illusions of Reform, one of the handiest refutations of the nonsensical claims of the Novus Ordo’s defenders.)
All that I will say here is that I guffawed when I read the final line: “The use of Latin…has indeed diminished…but it has not disappeared.” Give that bot an award for understatement! 99% of the places where you will find it are offering the traditional Latin Mass — that’s the main reason it hasn’t disappeared from the face of the earth!
Philosophical and Theological Objections
Now I wish to come to what seems to me the crux of the matter: Should anyone even be developing and deploying an AI chatbot like this one?
I’d like to begin with the typical line of defense you will hear almost everywhere. A Facebook correspondent wrote:
Between human flesh and mind on the one hand and the Revelation of the Divine Word there have always been technologies such as writing, printing, audio, visuals, and just now AI. Your argument is not a knockdown argument against the use of AI as you seem to imagine. AI may be able to be used for good ends — I don’t know and my natural scepticism leads me to be cautious about its use. Yet condemning AI out of hand in the current state of knowledge seems very high-handed to me.
In my opinion, those who defend AI programs like this, treating it as just another “forward step for technology” (and an inevitable one at that), are making a huge mistake.
AI is fundamentally different from all earlier forms of communication technology. A piece of papyrus, a pen, or a word processor does not talk back at you. It doesn’t answer your questions, structure your plot, create your characters, or guide your life.
All earlier forms of technology, while they do shape how we interact with nature and with each other, are essentially passive: they are tools that we use to accomplish our own aims. They are not tools that act upon us as if they were other humans. The pen is passive, and you must write with it. A novelist who blamed his bad novel on his pen would rightly be laughed at. The word processor (or the Substack editor) stares at you with a blank screen; you have to fill it with words.
An AI generator, on the other hand, is programmed to put out human-like content, actively, as we can see with “Fr. Justin.” He tries to sound like a person, indeed like a priest. In the greetings and farewells (which I have not quoted above), he calls you “my friend,” speaks of “our Catholic faith,” and says “may God bless you”; he constructs answers for you on the basis of which you are (presumably) expected to believe or to act.
This is not passive technology, but impersonating technology lining itself up alongside other rational agents. Defenders of AI show a singular blindness to the infinite difference there is between the two types of technology. It is not a quantitative threshold but a qualitative one.
Although it was egregious for Catholic Answers to create a “priest bot,” it will not be any less objectionable if the bot is a layman, a “catechist” as it were. For although there won’t be artificial absolutions, there will still be the illusion of a person actively answering on behalf of Christ and the Church. That is a mockery of both Christ and the Church. Leila Marie Lawler noted: “It isn’t about the thing being a priest or a layman, it’s about making something look like a human interaction that isn’t one. There’s a deception there.”
Beyond this more general point, we can look at the nature of what is being “discussed” with the digital non-person. You don’t put religious, spiritual, or theological questions to a computer database, no matter how “well stocked” it is — unless you think there’s only a difference in degree, not a difference in kind, between a living and breathing human person with an intellect and will, and a computer that spits out answers based on keywords, algorithms, patterns, and the like (shaped, no less, by random, democratic user interaction: “Justin will get better with time!”). As a friend of mine put it:
One of the major challenges with large language models like ChatGPT is that their goal is to produce an answer that seems true, not an answer that is true. There is obviously some overlap between those two goals, but they are not identical, much as sophistry and philosophy are not identical.
More bluntly, we can say there is no truth there, only appearance — a construct, not a concept; a Frankenstein of phrases stitched together and brought to life by electricity.
“Who” is a chatbot? I mean, whom are you speaking to? By whose authority does it teach? Has it got a charism of teaching? What are its credentials? Can it bear personal, living witness to the faith it pretends to teach? Who “owns” the output? If you complain to Catholic Answers about what Justin says, they could just shrug their shoulders and say “hey, don’t blame us, we just put your question in and spun the wheel!”
Of course, it’s only a matter of time before there’s a Calvinist chatbot, a Muslim chatbot, an Eastern Orthodox chatbot, and the rest of it. All of them will be churning out arguments on behalf of their positions. Shall we set them up for a public debate, as in the Middle Ages or the Reformation period? A new virtual War of Religion? To me all this seems like going deeper and deeper into Plato’s cave, further and further from the life of prayer, study, apprenticeship, discipleship, that we need in order to acquire a discriminating taste for the truth, or put differently, to bolster our sensus fidelium as Catholics.
I can see asking AI a mathematical question or something on botany or zoology, but should we ask it to evaluate great literature, to write our poetry, or to give us spiritual direction? Again, would you put AI in charge of a crisis situation in which many things are going wrong, others make no sense, still others are contradictions, and the whole is shot through with hypocrisy, duplicity, mendacity, and mayhem? I’m referring, of course, to the past sixty or so years in the Catholic Church. How is a bot supposed to navigate choppy waters like that?
It is one thing to have a simple catalog, encyclopedia, or Q&A catechism online, which people can search; for in that case, they are still using their intelligence and not trusting themselves to the “guru in the machine.” One might live with a “Catholic Source Bot” that merely said “Augustine wrote A, Aquinas argued B, Trent condemned C,” giving exact references to the literature, and noting which questions are disputed and which are nailed down (how about providing theological notes?). It’s quite a different thing to have a “live” program churning out constructed answers that have to synthesize 3,000 years of Scripture, Church Fathers, Doctors, mystics, magisterial texts, and theological interpretations, like a “robot rabbi” that somehow is expected to deliver the goods. This is profoundly inhuman and contrary to the way in which we are meant to seek for answers and discover them by the study of sources. The process of synthesizing, of deciding what to trust and what to discard, should never be left to a machine.
But even when it comes to providing a catalog of sources and citations, how do we know such an output is comprehensive, representative, well-chosen, appropriate, and so on? Selection and the weighing of sources will always be involved, and it is best that there be no illusion of objectivity and comprehensiveness — precisely the illusion an AI system conveys, with its quasi-authoritative, dispassionate, all-seeing “view from nowhere.”
Then there is the problem of the laziness — already an epidemic — that computer technology induces in us. Mortal man is so “outclassed” by the speed, power, and massiveness of such technology that he is strongly tempted to give up and let the machine do the “hard work” for him. Yes, I know something similar can be said about every new phase of technology (e.g., washing machines vs. scrubbing clothes on a board, or typewriters vs. pens), but as computing power increases exponentially, it brings with it qualitatively different problems. The more computers and the internet have invaded human life, the stupider and lazier people seem to be. All teachers know this to be true of their students if they have been in academia long enough to watch the generational patterns. And if we’re being honest, we’ll admit it’s true for ourselves. Who really wants to devote himself to the hard, long work of mastering languages, when you can get such a good quick translation with DeepL? And who’s going to bother putting years into the study of St. Thomas in order to develop a Thomistic mind, if instead someone offers you a “ThomasBot” that will formulate a perfect Thomistic response to every query, even providing sources?
Circling back to my original examples of “harmless AI,” can we not say there are problems even with some of those? For example, part of the fun of traveling is the serendipity: stumbling across towns or sites you hadn’t known about or planned to see, or unforseen opportunities that came to your attention because you didn’t have everything worked out to the last detail. Perhaps an AI travel agent would lessen this properly human adventure. Or if you had a grammar AI unit that corrected all your mistakes (I mean, doing a great job of it, unlike Microsoft Word’s grammar check), would you ever learn to write well? And more importantly, who is in charge of that virtual grammarian? Probably a cadre of politically correct programmers who will dumb down your language and eviscerate it of all “patriarchal” and “phallogocentric” remnants.
In short: when we outsource to the machines, we outsource something of our humanity, our freedom, our healthy uncertainty, our growth through work. We become diminished knowers and lovers, rather than the “supermen” that proponents of technology lure us into believing we will become (whether because, with incredible naivete, they believe their own gospel, or because, with incredible cynicism, they know it is a lie and can profit from it).
Let the Fr. Justins, Mr. Justins, Ms. Justines, and that whole digital tribe of unmen stay far away from the sacred realms of higher thought and Catholic life.
Indianapolis Lecture and Q&A now up on YouTube
As a brief roundup, let me just mention that the talk I gave last week, April 18, is now up at my YouTube channel:
Also posted is the 45-minute Q&A period, which I think you’ll enjoy.
Thank you for reading (and/or watching or listening), and may God bless you!
"...No bad things only bad uses of things.." I think Marshall McLuhan would disagree; in fact, I think he did, in "Understanding Media," iirc.
Excellent analysis, thank you.
Has anybody contacted Catholic Answers to ask about the controversial answers or to see if they offer a correction? If they are contacted and support the answers, then the AI bot by any name is created in their own image.