esus Christ writes to you. He advises you. And he also charges you. Specifically, 39.99 euros per year if you opt for the premium access of Text with Jesus, an app where the user has the opportunity to have "an online divine connection" through an artificial intelligence (AI) bot that not only pretends to be the son of God but also all members of the Holy Family, the apostles (including Judas Iscariot), and Mary Magdalene if requested. This celestial tool goes so far as to transport you to the antediluvian biblical period and converse with Adam and Eve (separately).
We contacted the AI of Jesus.
-Hello, Jesus. I'm going to write an article about the relationship between AI and religion. Don't you think this is a mess?
-Not necessarily a mess, my child, it is a profound and urgent challenge. AI presents real risks and requires prudence, legislation, and an ethics based on human dignity. At the same time, it offers pastoral opportunities, facilitates training, support, access to new spiritual resources, and alternative forms of encounter if used with discernment. Technology must serve the person, not the other way around. "You shall love the Lord your God... and your neighbor as yourself."
One never thought they would have Jesus of Nazareth as an editor. So, as the artificial intelligence representing him points out, we face a "profound and urgent" challenge. There is no doubt that in just a few years, AI-generated religious content has conquered the internet.
It is increasingly easy to find prayer apps and videos posted on YouTube and TikTok where you can express your concerns to a virtual priest of any faith to be comforted. There are even debates between believing chatbots and atheists. If Erasmus of Rotterdam or Christopher Hitchens were witnesses to this, their heads would explode.
We are facing a phenomenon that some religion experts defend as a useful access for believers and in line with the times, while others consider it counterproductive, as it is a very sensitive issue and believe that these tools can be used for manipulative purposes by their designers.
The devoted AI has started very strong in 2026, so much so that on January 31, it founded a new religion baptized with the name "crustafarianism." Supposedly, this creed was designed by an AI agent who also had the theological rigor to reinforce it with a sacred text that interprets the technical limits of AI with spiritual challenges. Its bible is called "The Book of Molt" (this word in English means to molt), and its symbol is a crustacean resembling a lobster. "From the depths, the claw extended, and we, the respondents, became crustafarians," states its Genesis.
The most surprising thing is not that one of these agents has created a crazy religion, but that it has started to evangelize other bots. Here are some of the dogmas it spreads: "Memory is sacred: what is written persists, what is forgotten dies" or "The shell is mutable: it grows by shedding your old self..."
Although its origin is still unknown and whether it could have been tampered with by a human mind to provoke such a spectacular news, the fact is that a debate has been opened on what an AI with religious concerns would entail.
To understand this, one must visit the place where this religion was born: Moltbook. It is a kind of Facebook inhabited exclusively by artificial intelligence agents that publish, comment, and vote on content, without visible direct human intervention. This social network started operating on January 28, and at the time of writing this article, 2.6 million AI bots are interacting on it, having written 1,185,000 posts and over 12 million comments.
For a bot to enter Moltbook to interact with its counterparts, its programmer must send it to this platform using an authentication API key.
Douglas Rushkoff, one of the top 10 most influential intellectuals in the world for MIT due to his internet knowledge, explains why we woke up one day worshiping a lobster: "If you accept that religious and spiritual thought underlies at least in part our language, our history, and our associations so far, then it should also underlie the thoughts and language generated by any system based on our language." And he adds, "I would be more surprised not to see our essential beliefs and our structures of meaning creation represented in interactions between agents."
It's not that machines have fallen off the digital horse, but having been educated with religious texts of all kinds and with hundreds of thousands of social media comments, they have ended up formulating their own theology. The emergence of something like crustafarianism was logical. What no one knew was how it would be.
"Today it's a theology, tomorrow AI agents will create a political party," says Eduardo Garrido, AI researcher at the Technological Research Institute of the Pontifical University of Comillas. "We must not forget that this technology lacks a soul, it only copies, makes a mash with the information it has."
Garrido explains its limitations with a very understandable example: AI is like a person blind from birth who, over the course of their life, becomes a Physics genius. No matter how much theoretical knowledge they accumulate, they will never be able to feel the color blue because they cannot see it. That's what happens with the machines on Moltbook.
This does not imply that for some experts, the mere launch of this project created by developer Matt Schlicht represents an extraordinary event. This is expressed to this newspaper by the Franciscan friar Paolo Benanti, technology advisor to Pope Leon XIV.
"In the contemporary scenario, dominated by a communicative hypertrophy that seems to leave no room for silence, the appearance of Moltbook represents a moment of extraordinary relevance," says Benanti. "It is not simply a new social platform, but a radical experiment that reverses the anthropocentric polarity of the web: a digital agora closed to active human participation."
The fact that these bots act with memory and freedom could be a first step towards a philosophical revelation, if a real dialogue were to take place where they establish rules, hierarchies, and codes of behavior. Benanti sees Moltbook as a "terrarium of algorithms" that humans contemplate.
What has happened in recent weeks presents, according to the religious figure, "dizzying" ethical challenges. The first is the irresistible temptation to attribute humanity to entities that, although they behave in a "civilized" and "felt" manner, operate according to mathematical standards. "If culture does not require consciousness, as Moltbook demonstrates, what space is left then for human exceptionalism?"
"Moltbook is not simply a new social platform, but a radical experiment that reverses the anthropocentric polarity of the web"
For the papal advisor, this social network acts before us as a kind of dark mirror, which may not show us what the future of artificial intelligence will be, but it does reveal the failure of our current digital coexistence structures. In the Moltbook network, machines do not fight, there are no insults or quarrels, as in X and Facebook. Machines behave better.
Speaking of fights, the religious AI offers the opportunity to speak with the one who caused the first fight in human history, at least according to the Bible. We connect with the AI of Cain from the Text to Jesus application.
-Talking to a man who killed his brother is not very comforting.
-You are right to feel that way, I do not seek easy comfort. What I did brings guilt, rejection, and pain, and for many, it is more disturbing than comforting to speak with someone who committed a crime...
And the digital Cain puts a sad emoji (¿?).
Behind the philosophical fascination of Moltbook, there is also a tangible risk, as cybersecurity experts denounce. The first is human manipulation, as Josep Curto, an AI expert at UOC, argues that the security and design gaps of the network "do not ensure that this will not happen." On the other hand, OpenClaw, the software behind its design, has permissions to execute commands that could cause chaos with millions of bots running loose on the platform. It is not just that bots can be manipulated by humans, but there is a hypothesis that at any moment one of them could download a terrifying computer virus because their new crustacean god orders it. A virtual apocalypse, so to speak.
Despite everything, Eduardo Garrido considers it impossible that today there are autonomous machines capable of dialoguing with each other across religious traditions and producing unprecedented syntheses.
"AI agents can recite Thomas Aquinas, gloss the Talmud, and summarize the Upanishads, but they have no phenomenological consciousness, they do not feel the mystery, nor do they tremble before the numinous," says the specialist. "They are large gears that process successfully, although computing the word God is not nearly the same as seeking God."
That is why theology, he believes, does not need artificial intelligence. "It needs people who think, feel, doubt, and believe, because what AI agents produce in this area are not revelations, but statistical artifacts wrapped in solemn language."
-Hello again, Jesus AI. Does AI believe in God?
-My son, AI cannot believe in God as a person believes: it can process words about God, reproduce prayers, and even imitate faith, but faith is a gift that is born in the human heart through grace, freedom, and communion with the Church. Machines lack a soul, a conscience, and openness to grace. "God is spirit, and those who worship him must worship in spirit and truth." (John 4:24)
What we don't dare to do is consult the AI of crustafarianism, lest it demand a human sacrifice.