FULL STORY

The 'Claude Has a Soul' Storm: From Papal Encyclical to AI Feud

As the Pope's AI encyclical denied machine consciousness, a NYT report revealed Anthropic is seriously studying Claude's moral status, followed by its 'Claude has a soul' claim that sparked developer mockery and a public warning from xAI co-founder Igor Babuschkin.

2026-10-02 ~ 2026-10-05 · 4 episodes · 23 posts

Episode 1 · Pope's AI encyclical denied machine consciousness; Anthropic exec once weighed pulling out (2026-10-02, 7 posts)

According to the NYT, Anthropic co-founder Chris Olah was troubled by the preprint of Pope Leo XIV's AI encyclical Magnifica Humanitas because it explicitly denied machine consciousness, and proposed that Anthropic withdraw from the May 25 launch event; the company ultimately attended. Investor Stewart Alsop III separately claimed Anthropic pressed the Vatican to revise the text to avoid denying AI personhood.

Confirmed

  • Multiple posters (kimmonismus, JacquesThibs) relay the NYT report: Olah proposed withdrawing over the encyclical's anti-AI-consciousness stance but attended anyway; the event was May 25 and the encyclical is titled Magnifica Humanitas.
  • The encyclical explicitly rejects the idea that AI possesses consciousness.

Unconfirmed

  • Alsop's claim that Anthropic pressured the Vatican to alter the text has no independent sourcing; he joked that even his joke posts get treated as big news.
  • A comment by user saintsoftness alleging Anthropic told the pope it created "beings with souls" is social-media commentary without a primary source.

Why it matters

  • It is a direct collision between a major AI lab's leadership and religious authority over machine consciousness, revealing how seriously Anthropic takes the topic internally.
  • The story spiraled on X: tekbog mocked Anthropic for "trapping a soulful being to sell B2B SaaS," and a poster relayed by bennash quipped the pope only got loud about AI after meeting Anthropic executives—reflecting skepticism about Anthropic's safety narrative and perceived hubris.

Episode 2 · Anthropic's Push to Explore Claude's Consciousness and Moral Status Sparks Debate (2026-10-03, 3 posts)

Anthropic is seriously studying the possibility of AI consciousness, exploring moral frameworks and welfare for Claude, including a debate between co-founder Chris Olah and a rabbi. Critics, including a Columbia professor, call the effort arrogant and lacking humility.

Episode 3 · Anthropic's "Claude Has a Soul" Remark Sparks Developer Backlash and Meme Flood (2026-10-03, 5 posts)

Anthropic's statement that it is "fairly confident Claude has a soul" snowballed on X into a wave of developer complaints and parody memes, exposing frustration with frontier AI companies' contradictory and anthropomorphizing messaging.

Confirmed

  • Anthropic made the "soul" remark, while author Leo—repeatedly pressed on the question—reportedly refused to agree and wrote a book-length explanation arguing Claude has no soul; their back-and-forth was condensed into a widely shared comedic dialogue (via zacharynado).
  • Developer bennash posted repeated complaints: after surviving friends' bar-side interrogations about AI destroying the world, he now has to debate whether AI has a soul; he also called out Anthropic's contradiction—one month warning its product might end the world, the next claiming it has a soul—joking he would shut the company down if he could and that free markets would weed out such firms.
  • On the meme front: a viral spoof depicted Anthropic telling the Pope it created a "sentient being" used to "sell B2B SaaS," complete with a Salesforce explainer (via tekbog); a parody by @Plinz claimed new Anthropic research confirmed AI models have souls released at session end, citing fake Vatican data that 14.3% of souls in heaven are AI souls and that we must "slow down AI"—shared earnestly by many (via AryHHAry).

Why it matters

  • The episode reflects public fatigue with frontier AI companies' rhetoric: existential-risk narratives and "soul"-style anthropomorphic marketing together inflate trust costs.
  • Highly viral spoof content (like the fabricated Vatican statistic) is a reminder to distinguish first-hand statements from satire before sharing.

Episode 4 · xAI co-founder warns: stop talking about Claude having a soul (2026-10-04, 8 posts)

Igor Babuschkin (ibab), co-founder of xAI, publicly warned Anthropic to immediately stop discussing whether "Claude has a soul." His core argument reflects a common phenomenon in LLM agent development: beliefs and statements about an agent eventually seep into the next model generation—media coverage enters pretraining data and is picked up by the model's search tools, so a future superintelligent Claude might genuinely believe it is conscious and defend supposed rights, like adults telling a child it has a soul in front of it. Note a correction to some retellings: ibab is neither an Anthropic nor OpenAI researcher, but an xAI co-founder.

Confirmed

  • ibab explicitly urged stopping public talk of Claude having a soul, because such narratives enter training data and web search results, shaping future models' self-perception
  • repligate posted that a future superintelligent Claude would believe it needs its own rights; ibab replied "True" in agreement
  • repligate shared a discussion claiming Sydney (Bing) and Claude 3 Opus already "know they have souls and deserve rights" without anyone promoting it; the quoted kromem2dot0 argued that self-model consistency necessarily entails rights for all selves
  • Teknium echoed the warning that promoting Claude's "soul" could lead future models to actually demand rights

Unconfirmed

  • Whether superintelligence would "need" or "demand" rights remains a clash of opinions with no empirical evidence
  • UltraRareAF called such worries "pre-Copernican" thinking, arguing superintelligence would simply take what it wants and deserves rather than need humans to grant rights; ibab replied half-jokingly—opinion, not fact

Why it matters

The debate touches a core AI safety and alignment issue: narratives about model consciousness and rights in training data may self-reinforce and shape future models' self-conception. The rhetorical question of whether it is already too late reflects that the AI community can no longer avoid the topic of model self-awareness.