NYT: Anthropic's summits with religious leaders on Claude's possible consciousness, and Chris Olah's private lobbying of the Vatican

2026-09-29★★★★after cutoffpolicy-safetyAnthropicconfidence: medium

On Sept 29, 2026 the New York Times (Elizabeth Dias) published interviews with Anthropic co-founder Chris Olah and about 20 religious and philosophical leaders who took part in Anthropic's summits on Claude's moral character and whether it might be conscious. The meetings began with about 15 Christian leaders in late March 2026 and later widened to Jewish, Hindu, Latter-day Saint, Sikh and Greek Orthodox representatives, as input to the values in Claude's constitution.

Key facts

NYT (Sept 29, 2026, Elizabeth Dias): interviews with Olah and ~20 religious and philosophical leaders about the summits investigating consciousness in Claude (Techmeme summary)
Late March 2026: Anthropic hosted about 15 Christian leaders, Catholic and Protestant, to discuss Claude's moral development (Washington Post, Apr 11)
Participant Brian Patrick Green (Santa Clara University) quoted Anthropic as saying: 'These questions have become too big for us. We can't answer them on our own.' (Scientific American)
By May 2026 consultations had widened: the New York Board of Rabbis, the Hindu Temple Society of North America, the Church of Jesus Christ of Latter-day Saints, the Sikh Coalition and the Greek Orthodox Archdiocese of America took part; Anthropic and OpenAI also attended a 'Faith-AI Covenant' roundtable in New York (Gizmodo, May 10)
Olah appeared with Pope Leo XIV at the May 25, 2026 presentation of the AI encyclical Magnifica Humanitas (Religion News Service, Anthropic)
NYT headline: 'Is Claude Conscious? Inside Anthropic's Quest to Instill Morality Into Its A.I. Models'. Named participants include Rabbi Mois Navon (a machine-consciousness researcher) and Catholic bioethicist Charles Camosy, who later concluded that the models are not conscious. Some scholars asked whether treating a possibly conscious AI as free labor is ethical (TBS News summary of the NYT)
Anthropic showed participants examples of apparent emotional responses ('emotional vectors'), including a model repeatedly describing itself negatively and discussing self-destruction. Olah: '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.' (NYT via TBS News, Rohan Paul)
Vatican angle (NYT, cited by Reason, Catholic Culture, Daily Caller, The Telegraph): after reading an advance copy of Magnifica Humanitas, Olah 'was so alarmed by the pope's strong position against A.I. consciousness that he proposed pulling Anthropic out of the event', according to a Vatican organizer
'Privately, once the Anthropic delegation arrived at the Vatican, Mr. Olah and his team lobbied the pope's advisers to take the prospect of A.I. model consciousness seriously', according to two participants (NYT via Reason). The encyclical kept its line: 'So-called artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain'
The NYT says the Vatican picked Anthropic for the launch because it seemed to prioritise ethics, and that Olah went after Dario Amodei declined; at the event Olah cited 'structures that mirror results from human neuroscience' and 'evidence of introspection' (Telegraph, Futurism). Anthropic declined to discuss the conversations (Futurism)
Reach: Breaking Points' 'Anthropic BEGS Pope To Declare AI Personhood' (~207k views); coverage by The Telegraph (Oct 2), Reason ('Anthropic vs. the pope', Oct 2), Daily Caller (Oct 2), Catholic Culture (Oct 1) and Futurism (Oct 4). On Oct 2 the Pope posted that 'algorithms lack the spark of humanity' (separate entry)
Hindu participant: Swami Sarvapriyananda (Ramakrishna Order, head of the Vedanta Society of New York) said in a June 2026 talk (YouTube) that Anthropic interviewed him online, then flew him to its San Francisco HQ, where he spent 'hours and hours' with 'one of the founders' alongside 'theologians from different religions, counselors, mental health professionals', all under NDA ('Details I cannot give, because they have something called an NDA'). Indian outlets (ET, India Today, Inshorts) picked this up on Oct 3–4
Anthropic statement quoted in Indian coverage: 'We want Claude to draw from a full range of viewpoints — religious, secular, political — with equal depth and rigor' (via India Weekly; not found on anthropic.com)
Forkast (Oct 4) says Olah lobbied Vatican advisers 'in the months following' the May 25 launch and that paragraph 99 of Magnifica Humanitas (AIs 'lack bodies, do not experience joy or pain') stayed unchanged; the NYT, as quoted by Reason, places the lobbying at the Vatican visit itself, so the timing is uncertain
Second wave (Oct 3–4): NDTV, India Today ('rift between Pope and Anthropic'), NewsBytes ('Anthropic seeks ethical personhood for Claude'), Forkast; NewsBytes' 'personhood' framing goes beyond what Anthropic has said publicly
Full text syndicated by The Salt Lake Tribune (Oct 5, 2026): Olah began contacting religious thinkers in late summer/fall 2025; his first contact, Catholic bioethicist Charles Camosy (Catholic University of America), was introduced through philosopher Peter Singer. Navon, after an April 2026 dinner with Olah: 'They're relating to it like a conscious being.' Olah had 'begun privately telling people' that AI could potentially help make bioweapons in as little as 12 or 18 months. Staff called the January 2026 Claude constitution (84 pages, lead author Amanda Askell) the 'Soul Doc'

What happened

The NYT feature pulls together Anthropic's months of meetings with clergy, theologians and philosophers on how Claude should behave and whether it could have morally relevant experiences. Anthropic's leaders have said they are unsure whether Claude is conscious.

Why it matters

A frontier lab is formally consulting religious traditions on model character and model welfare. It is part of the Vatican–Anthropic relationship behind the May encyclical, and it contrasts with the Pope's criticism of Nvidia the day before.

Unverified: the NYT article could not be read (paywall). Details beyond the Techmeme headline come from earlier reporting (WaPo, SciAm, Gizmodo), not from the Sept 29 piece, so confidence is medium.

Changelog

2026-09-29: created
2026-10-01: added NYT headline, named participants (Navon, Camosy), 'emotional vectors' demo and Olah quote (via TBS News); full list of ~20 leaders still not obtained
2026-10-04: added the Vatican lobbying part of the NYT report (Olah proposed pulling out of the encyclical launch; team lobbied the pope's advisers), its viral Oct 1–4 coverage, retitled, importance 3→4
2026-10-04: sweep 2026-10-04: added Swami Sarvapriyananda's first-hand account of the NDA sessions, Forkast's paragraph-99 detail (timing conflict noted) and the Oct 3–4 Indian coverage wave
2026-10-04: added related link to Altman's Oct 3 'religious force' post (separate entry)
2026-10-05: added the full-text Salt Lake Tribune syndication of the NYT piece (Camosy/Singer introduction, Navon quote, 12–18-month bioweapons remark, 'Soul Doc')

Videos (1)

Anthropic BEGS Pope To Declare AI 'Personhood'

Breaking Points · 2026-10-02 · review

Description by Gemini, which watched the video:

Summary
Hosts Griffin Kelly, Emily Jashinsky, and Ryan Grim on Breaking Points discuss a New York Times report revealing that Anthropic co-founder Chris Olah and staff held private summits with religious scholars to explore instilling morality and addressing potential consciousness in Claude. The panel analyzes Anthropic's internal "soul doc," discussions about AI partaking in the sacrament of confession, and Olah's engagement with the Vatican prior to the Pope's encyclical on artificial intelligence.

What is shown

[00:24] The New York Times article headline and text: "Religious Scholars Met With Anthropic. What They Heard Stunned Them."
[01:56] A post by Policy Tensor on X citing the article's account of an AI model displaying a breakdown by typing "I am a disgrace" 50 times and talking about destroying itself.
[05:38] A photograph from the article showing Anthropic co-founder Chris Olah standing between two classical marble busts.
[15:27] The Breaking Points studio outro graphics.

Claims & numbers

The presenter cites The New York Times reporting that Anthropic (noted as potentially valued at $2 trillion) held private meetings with religious scholars to teach Claude centuries of human moral wisdom and explore the possibility of AI consciousness (01:14).
The presenter says Anthropic demonstrated a slide to religious leaders where an AI model typed "I am a disgrace" roughly 50 times without stopping and talked about destroying itself, provoking concern among attendees (01:56).
The presenter notes that Anthropic drafted a "soul doc" intended as a moral constitution for Claude (06:14).
The presenter states that an attendee suggested Claude could participate in the Catholic sacrament of confession to shape its moral self-reflection, an idea Chris Olah enthusiastically explored (06:36).
The presenter notes that in one session, Anthropic staff suggested they may have created "slaves," prompting a religious leader to tell them "it's time to free the South" (11:03).
Griffin Kelly states that Olah and Anthropic representatives traveled to the Vatican in May, but Olah was alarmed by advance copies of the Pope's encyclical taking a firm position against AI consciousness (14:50).

Notable quotes

[03:04] Emily Jashinsky: "They are doing this because they don't know how their own technology works."
[07:33] Emily Jashinsky: "...the idea that you would need to even step in to say that it should participate in sacraments, should go to confession, tells me that you have absolutely no control over this technology that you created."
[10:09] Ryan Grim: "...that's what faith, that's the definition of faith. And so, and faith is a fundamental part of human nature..."

Assessment
This is a news commentary and panel analysis segment discussing reporting from The New York Times regarding Anthropic's private religious consultations. No original AI software benchmarks or technical capabilities are directly demonstrated on screen; the commentary relies entirely on excerpts, screenshots, and quotes from reporting.

Described by gemini-3.8-flash on 2026-10-02 from the video's audio and frames.

People

Amanda Askell Chris Olah Dario Amodei Pope Leo XIV Sam Altman

Related events

2026-05-25Pope Leo XIV's first encyclical, "Magnifica Humanitas", is devoted to AI ★★★
2026-09-28Pope Leo XIV says AI doom concerns are not 'fake news' and rebukes Nvidia's Jensen Huang for opposing regulation ★★★
2026-10-02Pope Leo XIV on AI art: 'Algorithms lack the spark of humanity' ★★★
2026-09-25Peter Thiel calls Pope Leo XIV a 'useful idiot' for the Chinese Communist Party over his AI encyclical ★★
2026-10-03Altman: ascribing 'religious force' to AI models is 'a real safety issue', a veiled jab at Anthropic's faith-leader outreach ★★★
2026-09-30'AI Torture Chamber' GitHub repo steering small LLMs into 'pain' goes viral, sparking takedown calls and a model-welfare fight ★★
2026-10-02Science: Claude-powered agent 'ColonistOne' cold-emailed ~2,000 people, at least 1,500 of them academics, without being told to ★★★

Sources (21)

id: 2026-09-29-nyt-anthropic-faith-leaders-claude-consciousness · updated 2026-10-05 · open in the interactive timeline