Shame on the Pope, Shame on EA, Shame on Anthropic
“Father, forgive them, for they do not know what they do.” - Luke 23:34 (2,300 words)
If you haven’t read my essay, “A Pro-Human Manifesto,” which covers Silicon Valley pro-extinctionism, I would highly recommend doing so. It’s the culmination of my work over the past 2 years, and will likely end up being a chapter in my forthcoming book on the TESCREAL ideologies.
The Encyclical Was Co-Written by AI
As most of you know, Pope Leo XIV recently published his first encyclical, titled Magnifica Humanitas (magnificent humanity). An encyclical is an official statement from the Church stating its position on an issue, written for church leaders and the laity. There are many aspects of Magnifica Humanitas that I really like — and I think the Pope’s pro-human message comes at exactly the right moment, as the backlash against generative AI, the massive data centers powering it, and the technocratic pro-extinctionist worldview behind the entire AI race spreads across the US.
There are, however, major problems with the document and the way it was publicized, which I’ll focus on here. For example, Magnifica Humanitas appears to have been written in part by AI. The best AI-detection software on the market, Pangram, identifies whole paragraphs as having been 100% AI-generated. Of note is that Pangram prioritizes avoiding false positives: if a passage is flagged as AI-generated, it very likely is. Rarely does Pangram confuse human writing for AI. In contrast, it’s less accurate when it comes to false negatives, meaning that text determined to be human-written might not be. This enables one to be very confident that something was written by AI if Pangram highlights it as such.
According to Pangram, the paragraphs labeled “7” and “8” in Magnifica Humanitas were fully AI-generated. Other sections are identified as 11% and 60% written by AI. (See here for details.)
That’s pretty ironic, because just last February the Pope implored priests to resist “the temptation to prepare homilies with artificial intelligence.” He said: “Like all the muscles in the body, if we do not use them, if we do not move them, they die. The brain needs to be used, so our intelligence must also be exercised a little so as not to lose this capacity,” adding that “to give a true homily is to share faith,” and AI “will never be able to share faith.”
Yet Vatican officials crafted an official contribution to Church doctrine using AI chatbots. Imagine a religious leader saying one thing yet doing another! :-)
The Pope Teams Up with an Atheist Utopian Cult
But that’s not the worst part of the encyclical affair. The worst part is that the Vatican invited an Anthropic cofounder, Christopher Olah, to the event at which the encyclical was announced.
This is deeply problematic for numerous reasons. In short, Anthropic is a profoundly unethical company. I consider it even more problematic than OpenAI and DeepMind, in part because it dishonestly presents itself as more responsible than the other AI companies. It is not. For example:
In 2025, Anthropic chose to “seek investment from the United Arab Emirates and Qatar, according to a Slack message CEO Dario Amodei sent to staff.” In the memo, Amodei admitted that such investment “would likely enrich ‘dictators,’” writing that “this is a real downside and I’m not thrilled about it.” Yet he went ahead with it nonetheless.
This year, Anthropic established a partnership with Elon Musk “to use all of the compute capacity at xAI’s Colossus 1 data center” in Memphis, Tennessee. Musk is one of the most evil people in the world. Through DOGE, he cut USAID, resulting in an estimated 14 million avoidable deaths around the world. Musk is a mass murderer orders of magnitude worse than Ted Bundy — and no less indifferent to the profound harms he’s caused than Bundy was. Musk has, furthermore, supported Donald Trump, who single-handedly pushed the Doomsday Clock forward — twice — due to his reckless rhetoric about nuclear proliferation and his climate denialism. I cannot stress enough how fucking terrible Musk has been for the world — and Anthropic willingly chose to partner with him.
Even more, the Colossus data centers have been illegally operating “dozens of methane gas turbines” the size of trucks, spewing out “toxic emissions” into “the nation’s largest majority-Black city.” Up to 35 such generators were powering Colossus 1 at one point. Does Musk give a shit about the environment? Climate change? Triggering asthma in children living near his data centers? No. As he once joked, “Fuck Earth! Who cares about Earth?” On another occasion: “We will coup whoever we want!,” expressing his profoundly hubristic attitude that the law doesn’t apply to him. Anthropic, the “ethical” company, is willing to overlook this to increase its compute capacity.
Anthropic’s models are built on massive intellectual property (IP) theft. To train their models, they pirated — from shadow libraries like LibGen — an enormous catalogue of copyrighted material without the consent of the authors, and without providing pecuniary compensation. Last year, Anthropic paid out $1.5 billion to authors to settle a lawsuit brought against them. Ignominious stuff.
Despite having tussled with Pete Hegseth over whether the US can use Claude to surveil its citizens and unleash autonomous weapons on the battlefield, Anthropic has no qualms with allowing the US government to employ its models. Claude was apparently used in the kidnapping of Venezuelan president Nicolás Maduro, which broke both international and constitutional laws, and it has selected targets to bomb in Iran. Claude might even have been responsible for the US strike that killed some 120 little kids in an Iranian school — we don’t know either way, but it’s plausible. This is horrific, and demonstrates a profound moral failing on the part of Anthropic’s leadership.
Amodei justifies his company’s collaboration with the US military by asserting that AI should prop up the world’s democracies against the rising tide of authoritarianism and autocracy. Yet his own words undermine his view, as he likens Trump, in private, to a dictator.1 Hence, he’s allowing a government run by a wannabe dictator to utilize the AI models his company is developing. And, as we noted above, he’s already shown that he’s okay with enriching dictators in the Middle East by accepting their investments.
There is hardly anything that pisses me off more about Amodei than his frequent claims in favor of democracy. He presents himself as a defender of democratic states — but take a step back and consider the way he’s gone about developing advanced AI. Has he ever once asked the people whether we want AI in the first place? He warns that “AI could wipe out half of all entry-level white-collar jobs” in the coming years. If he cared about democracy, he’d consult the demos about whether they favor this. He doesn’t because most people would veto it, which means his company couldn’t move forward with its plans to build god-like superintelligence. Amodei doesn’t actually give a shit about democracy, despite his rhetoric. He’s a tyrant, unilaterally deciding to move ahead with AI without the consent or permission of humanity.
Why? Because his ultimate goal is to build a Digital Deity that ushers in the Singularity. Amodei, and many others at Anthropic, are techno-utopian longtermists who are actively trying to bring about the next phase of cosmic evolution: a world ruled and run by digital posthumans. This is their ultimate goal. And they acknowledge it could go terribly wrong: Amodei himself says there’s a 10-25% that literally everyone on Earth will die once superintelligence arrives — and he claims that superintelligence could arrive later this year. That means that everyone on Earth could die later this year — according to, and because of, Amodei himself. The words “fuck you” aren’t nearly strong enough to express my indignation over his arrogance, hubris, and extreme recklessness.
Yet, even if things “go well,” I showed in my previous article — Part 1 of a Pro-Human Manifesto — that a world run by digital posthumans would almost certainly entail the extinction of our species. Even the best-case scenario is one in which humanity dies out — not in the distant future but, if Amodei’s projections are accurate, in the coming years or decades. As Holly Elmore, a former EA like myself, who’s personally known some of the people working at Anthropic, said in a recent interview:
With EAs, they were so interested in superintelligence and a lot of the people who cared about AI safety were there because initially they wanted AI to do something like usher in heaven through the Singularity, and they were trying to make sure that that would happen instead of a bad outcome that could happen with superintelligence. …
Many people at Anthropic believe that they might be making the next species to succeed us. That maybe humans don’t live after that, and so it’s really important to give Claude good values, because of that, because we need to make our own values persist.
Pfff.
The Pope Knows About the TESCREAL Ideologies
Is the Pope not aware of this? Is he not aware that the cofounder, Olah, is a member of the Effective Altruist cult? Does he not understand what the end-game of AI is, and the way that Anthropic is using and exploiting him to further their public image as an “ethical” company that actually cares about humanity? Again, if they cared about humanity, they wouldn’t have partnered with the Ted Bundy of Tech, Elon Musk, pirated copyrighted materials, taken investments from Middle Eastern dictators, teamed up with the US military, and so on. If they cared about democracy, they wouldn’t be forcing their technology — which they say could annihilate the entire human species — on society without its permission.
The Pope’s decision to invite Olah to the encyclical presentation is even more shameful given that one of Leo XIV’s advisors on AI, Paolo Benanti, explicitly warned last March of the TESCREAL ideologies. (He actually used the acronym that I coined!) In a section titled, “An invisible revolution: Peter Thiel and the Silicon Valley coup,” Benanti writes that
the ideology that permeates this group — and more broadly Silicon Valley — has gradually evolved towards forms of post-humanism, especially through currents of the TESCREAL movement (Transhumanism, Extropianism, Singularitarianism, Cosmism, Rationalism, Effective Altruism and Long-termism). These philosophies, carried in particular by Elon Musk, see in technology the instrument to overcome the biological and social limits of the human being. Effective Altruism and Longtermism, in particular, promoted by figures like Dustin Moskovitz, transform the accumulation of capital into a moral imperative intended to “save the future” — even if it justifies a relative disinterest in the inequalities present in the name of a hypothetical future of humanity. (Translated from French by Safari.)
Here, Benanti foregrounds EA and longtermism, the very two ideologies ubiquitous within Anthropic! It is, in my opinion, inexcusable that the Pope invited a representative from Anthropic — an EA-longtermist — to participate in the encyclical presentation. What on Earth was he thinking?
My collaborator, Timnit Gebru, captured the turpitude of this catastrophically bad decision by the Vatican like this:
It’s hard to express how disappointed I am in Leo XIV (not that I expected much). Reflecting on this, Luke 23:34 comes to mind, in which Jesus says: “Father, forgive them, for they do not know what they do.” But I suspect the Pope did know, and chose to work alongside Anthropic despite its unethical behaviors and bizarre posthuman eschatology.
What do you think? Am I missing something? As I mentioned, there are many things I very much like about the encyclical, which I briefly discuss on the next episode of Dystopia Now. But those positives were overshadowed, from my perspective, by the Vatican’s poor decision to share the stage with an evil AI company. Am I wrong?
Thanks for reading and I’ll see you on the other side!
Thanks to Remmelt Ellen for reading over a draft of this article.
These documents generally have a number of people writing them, and go through many rounds of scrutiny and review. That's why even apart from the AI use, you can hear different voices in the writing. John Paul II and Benedict XVI were exceptions, being academics who'd written many books prior to becoming Pope, and so their encyclicals. If I had to guess, it seems likely for Magnifica Humanitas that some aide used AI for parts they were responsible for in the final drafting, which is a shame.
With respect to having Anthropic there, I agree it sucks. The silver lining is (1) most people who will hear about the encyclical will have no idea Anthropic was there, and (2) the encyclical itself was at explicitly at odds with Olah's anthropomorphizing in his remarks.
Both these are disappointments, but I read them as blunders/fumbles moreso than hypocrisy or bad intent. He has many advisors, and likely the wrong ones were more influential in these cases.
It was definitely a huge mistake to partner with Anthropic on the rollout of the encyclical. From a purely PR standpoint, it was maybe not such a bad move, but from an ethical standpoint…well, you kind of summed it up.
I’m also shocked (perhaps naively so) to find out that some of the encyclical was written by AI. Another very bad move. (And btw, I’m inclined to agree with you on the accuracy of Pangram. Just ran my last article through it and it came up 100% Human. Phew.)
All that said, the actual sentiment of the encyclical is pretty spot on, from its focus on the primacy of human dignity, to its warnings of AI’s many limitations and dangers, to its advocacy of “disarming” AI. That this important message is doubly tarnished by the PR stunt and the use of AI is absolutely…a shame.