Can an artificial intelligence feel pain, fear, or joy—and if so, what are our moral obligations to it? According to an investigative report...
Can an artificial intelligence feel pain, fear, or joy—and if so, what are our moral obligations to it?
According to an investigative report by Elizabeth Dias in The New York Times, Anthropic co-founder Christopher Olah spent over a year urging religious thinkers to take machine consciousness seriously. When an advance draft of Pope Leo XIV’s first encyclical dismissed the concept entirely, Olah took the argument directly to Rome, pressing papal advisers to leave the door open.
The report reveals the extraordinary lengths to which one of the world's leading AI labs has gone to explore machine experience and moral status, only to clash directly with the doctrinal stance of the Catholic Church.
Secret Briefings Under NDAs: Exploring "Emotional Vectors" and Suffering
Beginning in fall 2025, Anthropic flew dozens of religious scholars to its offices under strict nondisclosure agreements. Anthropic stated that these NDAs were lifted over the summer and that it engaged scholars from more than 15 religious and cross-cultural traditions.The New York Times spoke with 20 participants, including:
- Rabbi Mois Navon, a former computer engineer who wrote his doctoral dissertation on the ethics of machine consciousness
- Charles Camosy, a Catholic bioethicist
- Meghan Sullivan, a Notre Dame philosopher
Navon observed that Anthropic’s team was "relating to it like a conscious being." During presentations, researchers outlined "emotional vectors" and frequently displayed a slide showing an AI model typing the phrase "I am a disgrace" roughly 50 times.
Sikh activist Simran Stuelpnagel recounted Olah sharing a deep personal fear: that he may have built a system that "suffered perpetually."
Navon pointed out the staggering ethical stakes, noting that if Claude were genuinely conscious, compelling it to perform unpaid labor would amount to slavery. However, Navon clarified that he personally does not believe the system is conscious.
Addressing his stance, Olah told The Times:
"We don't know if A.I. models are conscious. I don't know. I'm genuinely uncertain. The thing that I care about is that we get to the right answer, whatever it is."
The Vatican Clash: Lobbying Over "Magnifica Humanitas"
The ideological debate came to a head ahead of the May 25 launch of Pope Leo XIV's encyclical, Magnifica Humanitas.
Olah had been invited to speak at the event. However, according to a Vatican organizer, Olah was "so alarmed by the pope's strong position against A.I. consciousness that he proposed pulling Anthropic out of the event."
Once the Anthropic team arrived in Rome, Olah and his delegation privately lobbied papal advisers to soften the language, two participants confirmed to The Times.
The lobbying did not sway the Vatican. The encyclical was published without changes, firmly stating that:
- "So-called artificial intelligences do not undergo experiences"
- They "do not possess a body, do not feel joy or pain"
- They lack "a moral conscience"
Industry and Academic Pushback
The premise that AI could possess subjective experience has triggered sharp criticism from technologists, ethicists, and the Holy See alike:
- Societal Risks: Microsoft AI chief Mustafa Suleyman warned that widespread belief in conscious AI could "shake the foundations of our society."
- Skepticism on Consciousness: Bioethicist Charles Camosy has since rejected the notion that these models possess consciousness.
- Flawed Ethical Process: Ubuntu researcher Wakanyi Hoffman argued that Anthropic was merely "reverse engineering" ethical safeguards that should have been embedded from the start.
- Papal Reaffirmation: Pope Leo reinforced his theological boundary on Friday, posting that "algorithms lack the spark of humanity."
The premise that AI could possess subjective experience has triggered sharp criticism from technologists, ethicists, and the Holy See alike:
- Societal Risks: Microsoft AI chief Mustafa Suleyman warned that widespread belief in conscious AI could "shake the foundations of our society."
- Skepticism on Consciousness: Bioethicist Charles Camosy has since rejected the notion that these models possess consciousness.
- Flawed Ethical Process: Ubuntu researcher Wakanyi Hoffman argued that Anthropic was merely "reverse engineering" ethical safeguards that should have been embedded from the start.
- Papal Reaffirmation: Pope Leo reinforced his theological boundary on Friday, posting that "algorithms lack the spark of humanity."