Movemental Research Corpus

AI Adoption, Governance, and Formation in the Faith and Nonprofit Sectors

Version 1.0, May 8, 2026

---

Front matter

Purpose. Citation-quality reference document for use across Movemental's website, Field Guide PDF, investor materials, and customer-facing decks.

Status. Internal and external use approved. Every claim in this document has been verified against primary sources or explicitly tagged for revision.

Confidence tags.

Maintenance. This corpus should be re-verified every six months. Each section's primary sources are dated; sources older than 18 months should be flagged for re-confirmation. The next scheduled verification pass is November 2026.

Contact. Joshua Shepherd, CTO/Founder, Movemental.

---

Table of contents

1. Executive summary
2. AI adoption rates by sector
   • 2.1 Nonprofits
   • 2.2 Churches and pastors
   • 2.3 Cross-sector enterprise
   • 2.4 Higher education and theological schools
3. Governance and policy gaps
4. Documented harms
   • 4.1 FBI and federal sources
   • 4.2 Faith-sector specific incidents
   • 4.3 Nonprofit sector cybersecurity context
5. The capability gap
6. Adult learning and formation research
   • 6.1 Kolb's experiential learning theory
   • 6.2 Daniel Pink: autonomy, mastery, purpose
   • 6.3 Formation versus training (theological sources)
7. Sector spending and budget data
8. Competitive landscape
   • 8.1 Church management and giving
   • 8.2 Nonprofit CRM
   • 8.3 Faith-tech and values-aligned AI
9. Theological and philosophical grounding
10. Claims to drop or revise
11. Methodological caveats and use notes
12. Appendix: source index

---

1. Executive summary

The verification pass produced a corpus that is in most cases stronger than what Movemental has been citing, with a small number of revisions and one important correction.

What is verified. The Virtuous and Fundraising.AI 2026 Nonprofit AI Adoption Report holds up exactly as cited: 92% nonprofit adoption, 7% reporting major capability improvement, 81% individual use without shared workflows, 47% with no governance policy, n=346, released February 16, 2026. The Forvis Mazars February 2026 framing of AI governance as a board fiduciary responsibility is also verified, with a more precise reading available now. FBI IC3 data on AI-enabled fraud and voice cloning is verified for 2025 and is materially larger than most of the sector understands ($893M in AI-attributed losses in 2025 alone, plus a 400%+ jump in voice-cloning fraud cited by industry sources).

What needs correction. The "91% of church leaders use AI" claim attributed to Lifeway Research April 2026 is wrong as stated and needs to be replaced. Lifeway's actual September 2025 phone survey of 1,003 Protestant pastors (released April 21, 2026) found that 42% of pastors use AI for ministry (10% as regular users, 32% experimenting), with 56% non-users. The 91% figure does not appear in the Lifeway report. The closest comparable number comes from the Barna/Pushpay State of Church Tech 2026 report: 60% of church leaders personally use AI at least monthly, but only 33% report their church using AI in operations. The "9% have a formal AI policy" figure also needs correction: the actual number from Barna/Pushpay is 5% of churches with an established AI policy. The "25% of churches encountered AI scams" claim cannot be located in either Lifeway document and should be dropped or replaced with verified deepfake incident reporting.

What is new and stronger. Five sources should be added to the corpus immediately: the Barna/Pushpay 2026 State of Church Tech report (n=1,306 church leaders); Barna/Gloo State of the Church 2026 series on AI as spiritual authority (n=1,514 U.S. adults, November 2025); the Vatican's Antiqua et Nova (January 28, 2025) as a primary theological source; the BCG Widening AI Value Gap (September 2025, n=1,250) and McKinsey State of AI 2025 (n=1,993) for the cross-sector adoption-to-value gap; and MIT NANDA The GenAI Divide (July 2025), which documents a 95% pilot failure rate. Together these sources turn Movemental's "adoption without capability gain" thesis from a single-study claim into a converged finding across nonprofit, enterprise, and faith-sector research.

The strategic implication. The data is stronger for Movemental's argument than the existing pitch suggests. The pattern is now documented across at least four independent 2025–2026 studies: widespread adoption, a 5–7% high-performer cohort, and a governance gap that consistently predicts who lands in the high-performer cohort. The "formation precedes deployment" thesis sits well inside the empirical literature.

---

2. AI adoption rates by sector

2.1 Nonprofits

Source 2.1.1: Virtuous & Fundraising.AI, The 2026 Nonprofit AI Adoption Report | February 16, 2026 | VERIFIED

Key findings.

• 92% of nonprofits use AI in some capacity
• 79% report small to moderate efficiency gains
• 7% report major improvements in organizational capability
• 81% use AI individually without shared workflows
• 47% have no AI governance policy
• 4% have documented, repeatable AI workflows
• 65% characterize their AI use as reactive and individual
• 18% operational across team workflows
• 7% embedded into goals, budgets, performance indicators
• Top barriers among regular users: privacy and security concerns (32%), time and capacity (31%), staff skepticism (19%)

Methodology. Benchmark survey of 346 nonprofit organizations, late 2025.

URLs.

• https://virtuous.org/resource/the-2026-nonprofit-ai-adoption-report-download/
• https://www.prnewswire.com/news-releases/virtuous-and-fundraisingai-release-2026-nonprofit-ai-adoption-report-revealing-gap-between-ai-use-and-real-fundraising-impact-302688325.html

Source 2.1.2: TechSoup & Tapp Network, State of AI in Nonprofits 2025 | January 24, 2025 | NEW

Key findings.

• 96% of surveyed nonprofits report at least basic understanding of AI
• 85.6% are exploring AI tools
• 24% have a formal AI strategy (76% do not)
• Larger nonprofits (>$1M annual budget) adopt AI at roughly twice the rate of smaller organizations (66% vs 34%)
• Only 1% actively oppose AI adoption

Methodology. Survey of 1,321 nonprofit professionals, 2024.

URL. https://page.techsoup.org/ai-benchmark-report-2025

2.2 Churches and pastors

Source 2.2.1: Lifeway Research, Pastors' Views on Artificial Intelligence | Survey conducted September 2–24, 2025; released April 21, 2026 | VERIFIED (corrects Movemental's existing claim)

Key findings.

• 10% of U.S. Protestant pastors are regular AI users
• 32% are experimenting with AI
• 18% are waiting to see better examples
• 18% are intentionally avoiding AI
• 20% are ignoring AI
• 3% are not sure
• Total pastor AI use for ministry: 42%; non-users: 56%

Concerns.

• 84% worry AI-generated content must be edited for errors
• 81% find it hard to ensure reliable sources
• 76% concerned about programming bias
• 62% concerned about non-disclosure
• 59% concerned about plagiarism
• 55% say "God has always shared his Word through people, and AI isn't a person"

Demographic patterns.

• Pastors 18–44 (40%) and 45–54 (37%) more likely to experiment than 65+ (23%)
• Urban regular users (11%) vs rural (5%)
• Master's (10%) and doctoral (14%) degree holders are more likely regular users than no-college (5%)
• Congregations of 250+ are most likely to experiment (43%)
• Congregations under 50 are most likely to ignore (28%)

Denominational.

• Holiness most likely to adopt (43% experimenting, 18% regular)
• Baptists and Lutherans most likely to ignore (22% each)

Methodology. Stratified random sample, phone survey of 1,003 Protestant senior pastors, weighted by region and church size, ±3.3% margin of error.

URLs.

• https://research.lifeway.com/wp-content/uploads/2026/04/Lifeway-Pastors-Sept-2025-AI-Report.pdf
• https://research.lifeway.com/2026/04/21/pastors-churchgoers-see-ai-as-concerning-and-confusing/

Source 2.2.2: Barna Group & Pushpay, Technology for Missional Impact: State of Church Tech 2026 | Released March 9, 2026 | NEW (should replace Movemental's "91% / 9% policy" pairing)

Key findings.

• 60% of church leaders personally use AI at least once a month
• 24% never use AI
• 33% report their church is currently using AI in any part of ministry or operations
• 58% report their church is not using AI; 8% unsure
• 64% of leaders believe it is important for a church to have an AI use policy (24% extremely, 40% somewhat)
• Only 5% of churches have an established AI policy
• 19% of leaders confident their church's data is ready to produce meaningful AI insights

Concerns.

• 83% concerned about data privacy
• 51% "very concerned" about plagiarism and message integrity
• 49% "very concerned" about loss of authenticity in preaching
• 65% worry AI could displace their spiritual guidance
• 70% worry AI could erode congregants' trust

Optimism.

• 95% agree technology opens new opportunities for ministry
• 94% agree it helps the church fulfill its mission
• 79% say technology has improved congregational connection

Methodology. Survey of 1,306 U.S. church leaders, November 11 – December 8, 2025; combined Pushpay (644) and Barna (661) networks; 1,027 completed all questions.

URLs.

• https://www.barna.com/research/church-leaders-ai-usage-concerns/
• https://www.globenewswire.com/news-release/2026/03/09/3251986/0/en/

Source 2.2.3: Barna Group & Gloo, State of the Church 2026: AI Series | Released February 2026 onward | NEW

Key findings.

• Nearly 1 in 3 U.S. adults agree spiritual advice from AI is as trustworthy as advice from a pastor; 2 in 5 among Gen Z and Millennials
• 62% of practicing Christians say AI is making the world a better place (vs. 53% of U.S. adults overall)
• 66% of practicing Christians say AI is improving their lives
• 57% of practicing Christians say AI is a threat (matches national average)
• 53% say AI creates more problems than it solves
• 54% say AI is biased
• 19% of practicing Christians sit in the "high opportunity / high risk" quadrant, vs. 10% of non-practicing Christians and 7% of Americans with no faith
• 31% of practicing Christians say they want guidance from pastors on how to navigate AI; only 12% of pastors say they feel comfortable teaching about it
• ~40% of practicing Christians say AI has helped them with prayer, Bible study, or spiritual growth
• 41% of pastors report using AI for Bible study

Methodology. U.S. adult survey n=1,514, online, November 2025, representative quotas.

URLs.

• https://www.barna.com/research/state-of-the-church-2026-trends/
• https://www.barna.com/research/christians-view-ai-gift-threat/

Source 2.2.4: Lifeway Research, Protestant Churchgoer Views on Artificial Intelligence | Survey conducted September 2–7, 2025; released April 21, 2026 | NEW

Key findings.

• 44% see nothing wrong with their pastor using AI for sermon preparation
• 43% disagree (24% strongly)
• 13% not sure
• 42% would value hearing a sermon applying biblical principles to AI
• 61% are concerned about AI's influence on Christianity

Methodology. Online panel survey of 1,200 American Protestant churchgoers (attend at least monthly), ±3.2% margin of error.

URL. https://research.lifeway.com/wp-content/uploads/2026/04/American-Churchgoers-Sept-2025-AI-Report.pdf

2.3 Cross-sector enterprise

Source 2.3.1: McKinsey & Company, The State of AI in 2025: Agents, Innovation, and Transformation | November 2025 | NEW

Key findings.

• 88% of organizations use AI in at least one business function (up from 78% in 2024)
• ~33% report scaling AI across the enterprise; ~67% still in experimentation or pilot mode
• 39% attribute any EBIT impact to AI; most of those say <5% of EBIT is attributable to AI
• 6% of respondents qualify as AI high performers (>5% EBIT impact and significant value)
• 62% experimenting with AI agents; 23% scaling agents in at least one function
• High performers are 3.6x more likely than peers to pursue transformative change
• 55% of high performers have fundamentally redesigned workflows (vs. ~20% sector-wide)

Methodology. Online survey, June 25 – July 29, 2025; 1,993 respondents across 105 nations.

URL. https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai

Source 2.3.2: Boston Consulting Group, The Widening AI Value Gap: Build for the Future 2025 | September 2025 | NEW

Key findings.

• 60% of companies generate no material value from AI investment (up from prior year)
• 35% scaling and seeing some returns
• 5% are "future-built" companies creating substantial value at scale

Future-built company performance vs. peers.

• 1.7x revenue growth
• 3.6x three-year TSR
• 2.7x return on invested capital
• 1.6x EBIT margin
• 3.5x patents

Methodology. Global study of 1,250 firms.

URL. https://media-publications.bcg.com/The-Widening-AI-Value-Gap-October-2025.pdf

Source 2.3.3: MIT NANDA Initiative, The GenAI Divide: State of AI in Business 2025 | July 2025 | NEW

Key findings.

• 95% of enterprise GenAI pilots fail to generate measurable financial returns
• 5% of integrated AI pilots achieve rapid revenue acceleration
• 80%+ of organizations have piloted ChatGPT/Copilot; ~40% report deployment; only 5% of custom enterprise AI tools reach production
• $30–40B in enterprise GenAI spend with most yielding no business return

Methodology. 150 executive interviews, surveys of 350 employees, analysis of 300 public AI deployments.

URL. https://fortune.com/2025/08/18/mit-report-95-percent-generative-ai-pilots-at-companies-failing-cfo/

2.4 Higher education and theological schools

Source 2.4.1: Stanford HAI, AI Index Report 2025 | April 2025 | NEW

Key findings.

• 78% of organizations using AI in 2024 (up from 55% in 2023)
• 71% of those using GenAI in at least one business function (up from 33% in 2023)
• AI-related incidents reached 233 in 2024, a 56.4% increase over 2023
• $109B U.S. private AI investment in 2024

URL. https://hai.stanford.edu/ai-index/2025-ai-index-report

Source 2.4.2: Stanford HAI, AI Index Report 2026 | Released April/May 2026 | NEW

Key findings.

• Organizational AI adoption reached 88%
• ~80% of university students now use generative AI
• Industry produced over 90% of notable frontier models in 2025

URL. https://hai.stanford.edu/ai-index/2026-ai-index-report

Source 2.4.3: Association of Theological Schools, Atla, In Trust Center: F[AI]thfully Co-Creating series | 2024–2026 | QUALIFIED

Key findings.

• ATS does not yet have a formal accreditation standard requiring AI policies
• The current 2025 Standards Review (initiated February 2024) does not include a specific AI standard
• Movemental's claim that there is a "near-universal absence of an Association of Theological Schools standard" on AI is directionally correct: ATS is convening conversations and offering microcredentials but has not promulgated a binding standard
• The F[AI]thfully Co-Creating program (ATS, Atla, In Trust) launched 2024 and now offers three microcredentials including "AI Policy, Pedagogy, and Design in Theological Education" (announced August 2025)
• Individual seminaries have policies. Reformed Theological Seminary published an AI Acceptable Use Policy in its Global Student Handbook (2025). Saint Meinrad Seminary published its AI Policy with three instructor models (Strict, Moderate, Permissive)
• Sample academic study: Asia-Pacific Nazarene Theological Seminary (Del Rosario & Muan, Asia Journal of Theology, October 2025) found significant adoption among 39 students and 15 faculty

URLs.

• https://www.atla.com/ai-te-series/
• https://www.ats.edu/StandardsReview2025

Note on "near-universal student AI use." Movemental should cite the Stanford HAI 2026 figure (~80% of university students use generative AI), not a theological-school-specific number, because no comprehensive ATS-wide student-use survey has been published as of May 2026.

---

3. Governance and policy gaps

Source 3.1: Forvis Mazars, AI Governance for Nonprofit Boards | February 13, 2026 | VERIFIED

Authored by Karen Bodach and Dan Prater. Frames AI governance as a fiduciary expansion of the board's traditional duties of Care, Loyalty, and Obedience.

Recommendations.

• Resist the "efficiency trap"
• Address algorithmic blind spots through bias audits
• Codify responsibility using a framework such as NIST AI RMF
• Recruit board members with technology and data governance experience
• Require vendor accountability
• Incorporate AI risk into enterprise risk management

Cites NPQ ("How Nonprofits Can Resist the AI Efficiency Trap," October 28, 2025) and Council of Nonprofits ("Board Roles and Responsibilities," 2025) as supporting governance literature.

URL. https://www.forvismazars.us/forsights/2026/02/ai-governance-for-nonprofit-boards

Source 3.2: Forvis Mazars, AI Strategy: A Road Map From Readiness to Implementation | March 2026 | NEW supporting source

Key findings.

• Cites Financial Executives International 2026 report: 88% of organizations regularly use AI in at least one business function
• Notes "major" AI concerns dropped from 50% in 2024 to 36%, suggesting maturing risk practices

URL. https://www.forvismazars.us/forsights/2026/03/ai-strategy-a-road-map-from-readiness-to-implementation

Methodological note on the "47–76% no-governance" range. Both endpoints are real but they come from different surveys.

• 47% is the Virtuous/Fundraising.AI 2026 figure for nonprofits with no AI governance policy (n=346)
• 76% is the TechSoup 2025 figure for nonprofits without a formal AI strategy (n=1,321), distinct from a governance policy

These should be cited side-by-side as a range of governance gaps, not as a single metric.

Methodological note on "81% individual staff AI use without organizational policy." The precise Virtuous wording is: "81% of organizations report using AI individually and on an ad hoc basis." Use this wording.

Source 3.3: ECFA (Evangelical Council for Financial Accountability) | Standards last formally revised pre-AI | NEW context

Key context.

• ECFA accredits ~2,700 Christian nonprofit organizations against Seven Standards of Responsible Stewardship
• Founded 1979; standards focus on doctrinal integrity, board governance (minimum 5 members, majority independent), audited financials, transparency, fundraising integrity, conflicts of interest
• As of May 2026, ECFA has not published a binding AI standard or accreditation requirement
• This is a real gap and an opportunity for Movemental to position alongside or ahead of ECFA action

URL. https://www.ecfa.org/About.aspx

---

4. Documented harms

4.1 FBI and federal sources

Source 4.1.1: FBI Internet Crime Complaint Center, 2025 IC3 Annual Report | Published 2026 | VERIFIED

Key findings.

• For the first time, IC3 broke out AI-enabled fraud as a dedicated category
• 22,364 complaints with an AI-related descriptor in 2025
• $893 million in adjusted losses tied to AI-related scams
• $352 million of those losses sustained by Americans 60+
• Total 2025 cybercrime losses exceeded $20.877 billion across more than 1 million complaints
• BEC (Business Email Compromise) losses: $3.046 billion in 2025
• AI-attributed BEC losses: more than $30 million
• AI-employment fraud (deepfake interviews): ~$13 million in losses
• AI-romance scams: $19 million confirmed AI-nexus losses
• Voice cloning is now layered into BEC attacks for follow-up "CFO/CEO" calls reinforcing wire transfer requests

URLs.

• https://www.secureworld.io/industry-news/ai-enabled-fraud-topped-893m-fbi
• https://www.aarp.org/money/scams-fraud/fbi-ftc-report-2025-losses/

Source 4.1.2: FBI Public Service Announcements | 2024–2025 | VERIFIED

• Criminals Use Generative Artificial Intelligence to Facilitate Financial Fraud, December 3, 2024 (Alert I-120324-PSA): documents voice cloning, AI-generated video, and synthetic content in fraud schemes; recommends family code-words and verification protocols. https://www.ic3.gov/PSA/2024/PSA241203
• Senior US Officials Impersonated in Malicious Messaging Campaign, May 15, 2025 (PSA250515): AI-generated voice messages and texts (vishing/smishing) targeting current and former senior officials and their contacts since at least 2023. https://www.ic3.gov/PSA/2025/PSA250515
• Senior U.S. Officials Continue to be Impersonated, December 19, 2025 (PSA251219): update; activity continued. https://www.ic3.gov/PSA/2025/PSA251219
• Industry tracking cited by BlackFog: voice cloning fraud increased 400%+ in 2025

4.2 Faith-sector specific incidents

Source 4.2.1: WIRED reporting (November 2025) and follow-up coverage | NEW

Documented incidents.

• AI deepfake impersonation of Catholic priest Father Mike Schmitz (1.2M YouTube subscribers)
• Pastors in Alabama, Florida, Missouri, Nebraska, New York, and a Philippines megachurch issued public warnings about AI impersonation scams targeting their congregations
• Pastor Alan Beauchamp (Ozarks): Facebook account hacked; AI-generated cryptocurrency trading certificate posted with his likeness urging congregants to participate
• Evangelical church in Nebraska issued AI "scammer alert" after texts purported to come from one of their pastors
• Documented in AI Incident Database as Incident 1355: https://incidentdatabase.ai/cite/1355/

Coverage. WIRED (November 2025), VICE (January 2026), Gizmodo (December 2025), ASIS Security Management (January 2026), KnowBe4 CyberheistNews.

Source 4.2.2: Catholic Church / Vatican | NEW

Documented incidents.

• Vatican Dicastery for Communication reports receiving "dozens" of deepfake reports per day involving Pope Leo XIV
• Father Rafael Capó (St. Thomas University, Miami) impersonated by AI accounts soliciting donations
• Bishop Robert E. Barron (Word on Fire), Cardinal Marc Ouellet, Father Mike Schmitz among other documented Catholic deepfake victims
• AI-generated bishop "confronting ICE agents" video circulated on Instagram, Facebook, TikTok with viewers fooled into believing it was real
• Pope Leo XIV's January 24 message for the 60th World Day of Social Communications addressed deepfake and digital fraud risks

URL. https://www.osvnews.com/analysis-deepfake-popes-and-bishops-abound-heres-how-church-can-push-back-ai-attack-on-truth/

Source 4.2.3: TD Jakes deepfake / disinformation incident | July 2025 | NEW

Pastor T.D. Jakes had to address AI-generated rumors connecting him to Sean Combs sex parties. Investigation found 44,000 bot accounts spreading the claims.

Source. ASIS Security Management, January 2026.

4.3 Nonprofit sector cybersecurity context

Source 4.3.1: Identity Theft Resource Center and supporting sources | NEW

Key findings.

• Cyberattacks targeting nonprofits increased 50% between 2020 and 2024
• 2020 ransomware attack on Blackbaud exposed donor and constituent data at >13,000 organizations; >$49.5M multistate settlement plus $6.75M California settlement (2025)
• BDO 2025 nonprofit cybersecurity reporting: 30% year-over-year increase in weekly cyberattacks against nonprofits in 2024; 78% of organizations feel cyber resilience is insufficient; average data breach cost reaches $2M; 68% of breaches involve a human element
• Give.org Donor Trust Report: ~69% of donors worry their data could be hacked; 80% would stop or hold off giving after learning of a breach

URLs.

• https://www.mgocpa.com/perspective/crucial-role-of-cybersecurity-for-nonprofit-organizations/
• https://stratuslive.com/blog/nonprofit-donor-data-security-guide/

---

5. The capability gap

The pattern of widespread adoption with limited capability gain is now documented across at least six independent 2025–2026 studies, converging on a high-performer cohort of roughly 5–7%.

What predicts the high-performer cohort. McKinsey identifies workflow redesign as the single strongest correlate of AI EBIT impact: high performers are 3.6x more likely to pursue transformative change and 55% have fundamentally redesigned at least some workflows, vs. roughly 20% sector-wide. BCG's "future-built" cohort is differentiated by an AI-first operating model, business-IT co-responsibility, and modular architecture. Virtuous identifies the same pattern in nonprofits: organizations with shared workflows and outcome measurement are the ones reporting major capability gains, while 65% are stuck in reactive individual use.

The translation for missional organizations. The data supports the claim that the gap between adoption and capability is primarily an organizational, formation, and workflow problem rather than a tools problem. This is the strongest empirical foundation available for Movemental's "formation precedes deployment" thesis.

---

6. Adult learning and formation research

6.1 Kolb's experiential learning theory

Source 6.1.1: David A. Kolb, Experiential Learning: Experience as the Source of Learning and Development | Original 1984; second edition 2014 | VERIFIED foundational source

Four-stage cycle: Concrete Experience → Reflective Observation → Abstract Conceptualization → Active Experimentation. Foundational text for "learning by doing" framing in adult and organizational learning.

Source 6.1.2: Bergsteiner, Avery, & Neumann, Studies in Continuing Education | 2010 | NEW

Critique from a modeling perspective; identifies graphic syntax errors and definitional issues. URL: https://www.tandfonline.com/doi/abs/10.1080/01580370903534355

Source 6.1.3: Morris, T. H., Interactive Learning Environments 28(8) | 2019/2020 | NEW

Systematic review and revision: proposes "experiential learning consists of contextually rich concrete experience, critical reflective observation, contextual-specific abstract conceptualization, and pragmatic active experimentation." URL: https://www.tandfonline.com/doi/abs/10.1080/10494820.2019.1570279

Source 6.1.4: Dillard, Sisco, & Collins, New Directions for Adult and Continuing Education | 2024 | NEW

Argues for an expanded ELT framework integrating technology, interdisciplinarity, and cultural responsiveness. URL: https://onlinelibrary.wiley.com/doi/full/10.1002/ace.20539

Recommended Movemental citation. "Kolb's experiential learning theory (1984), updated by Morris (2019) and Dillard et al. (2024), grounds the cohort-based, learn-by-doing methodology."

6.2 Daniel Pink: autonomy, mastery, purpose

Source 6.2.1: Daniel H. Pink, Drive: The Surprising Truth About What Motivates Us | Riverhead Books, 2009 | VERIFIED

Three intrinsic motivators.

• Autonomy: self-direction over task, time, technique, team
• Mastery: continual improvement at something that matters
• Purpose: contributing to something larger than self

Draws on Self-Determination Theory (Deci & Ryan), Harlow's primate motivation studies, and MIT/Federal Reserve research on incentives and performance on cognitive tasks.

Pink's 2009 TED Talk ("The puzzle of motivation," TEDGlobal 2009) preceded the book and establishes the core framework.

6.3 Formation versus training (theological sources)

Source 6.3.1: James K.A. Smith, Desiring the Kingdom (2009) and You Are What You Love (2016) | Baker Academic / Brazos Press | NEW

Argues humans are primarily "desiring agents" rather than "thinking things." Liturgies (repeated practices loaded with vision of the good life) form character and desire more than cognitive content. Distinction between information (training) and formation (re-ordering of loves through embodied practice). Direct theological grounding for Movemental's "formation precedes deployment" thesis.

Source 6.3.2: Vatican, Antiqua et Nova: Note on the Relationship Between Artificial Intelligence and Human Intelligence | Dicastery for the Doctrine of the Faith and Dicastery for Culture and Education, January 28, 2025; approved by Pope Francis January 14, 2025 | NEW

117-paragraph note treating the anthropological and ethical implications of AI.

Key paragraphs.

• ¶67: "Current approaches to the technology can paradoxically deskill workers, subject them to automated surveillance, and relegate them to rigid and repetitive tasks"
• ¶105: warns of humanity becoming "enslaved to its own work"
• ¶112: AI "should be used only as a tool to complement human intelligence rather than replace its richness"

URL. https://www.vatican.va/roman_curia/congregations/cfaith/documents/rc_ddf_doc_20250128_antiqua-et-nova_en.html

Source 6.3.3: University of Notre Dame DELTA Framework | Launched September 2025; supported by $50.8M Lilly Endowment grant (December 2025) | NEW

Five faith-informed categories for AI ethics: Dignity, Embodiment, Love, Transcendence, Agency.

URL. https://ethics.nd.edu/programs/faith-based-frameworks-for-ai-ethics/delta/

Source 6.3.4: AI and Faith (aiandfaith.org) | Pluralist nonprofit (Seattle) | NEW

Connects technologists, theologians, ethicists across major religions. Partnership with Institute for Security and Technology (IST) and Future of Life Institute on the "Religious Voices and Responsible AI" initiative.

Source 6.3.5: Christianity and AI Conference | Harvard CCG, 2025 | NEW

Praxis (Andy Crouch as chair of one panel), MIT Media Lab (Rosalind Picard), Gordon-Conwell, Oxford. URL: https://ccg.fas.harvard.edu/conference-pages/christianity-and-ai

---

7. Sector spending and budget data

Direct nonprofit and church AI spending data is sparse in the public record. The strongest available signals follow.

Cross-sector benchmarks. Gartner forecast: ~$1.5T total AI spending in 2025. MIT NANDA: $30–40B in enterprise GenAI spend in 2024–2025 alone. McKinsey high-performer cohort allocates >20% of digital budget to AI initiatives.

Foundation and grant programs targeting nonprofits.

• OpenAI People-First AI Fund ($50M total): $40.5M distributed by December 2025 to 208 U.S. nonprofits via initial open application; remaining $9.5M for board-directed grants
• Google.org Accelerator: Generative AI ($30M global, 2025 cohort): 20 nonprofits selected, six months of structured support, Google Cloud credits, pro bono technical assistance. Google.org also committed $10M to Project Evident, Tech:NYC Foundation, and community foundations for AI workforce training (March 2025); part of $75M AI Opportunity Fund
• Patrick J. McGovern Foundation: $75.8M across 149 grants in 2025 ($500K–$750K typical), invitation-based, focused on AI in climate, human rights, health equity, crisis response across 13 countries
• Microsoft AI for Good Open Call: $5M in Azure credits, currently restricted to Washington State; Microsoft for Nonprofits provides $2,000 in annual Azure credits and discounted M365 nonprofit licenses (Microsoft 365 Copilot now at 15% discount for eligible nonprofits)
• AWS Imagine Grant Program: tiered award structure for nonprofits

---

8. Competitive landscape

8.1 Church management and giving

Source 8.1.1: Pushpay | VERIFIED

• Acquired by Sixth Street Partners and BGH Capital, completed May 19, 2023, for ~NZ$1.3B (NZ$1.34/share); deal announced October 28, 2022
• Now private, headquartered in Auckland with U.S. operations
• 2026 strategic activity: launched 2026 State of Church Technology report with Barna (March 2026); acquired Nurture.io for pastoral care/engagement intelligence (April 22, 2026); launched Pushpay Goes grant program for church planters (March 2026)
• Reportedly raised $1.26B in cumulative capital across history (PitchBook); ~517 employees
• AI positioning: launched AI Policy Generator (March 2026) to fill the 5%-policy gap their own research identified

URLs.

• https://www.willkie.com/news/2022/11/willkie-advises-sixth-street-and-bgh-capital-in-acquisition-of-pushpay
• https://pitchbook.com/profiles/company/92985-76

Source 8.1.2: Subsplash | VERIFIED

• Acquired by Roper Technologies on July 20, 2025, for $800M; K1 Investment Management exited
• December 2024: acquired Pulpit AI ("instantly create clips, discipleship material, social posts, and more from every sermon")
• AI positioning: integrated Pulpit AI into Subsplash Platform; partnered with Southeastern University (September 2025) for student/Pulpit AI deployment

URL. https://tracxn.com/d/companies/subsplash/

Source 8.1.3: Tithe.ly | NEW

• Acquired Elvanto, expanded into website builder and mobile apps via acquisition strategy
• Last institutional funding round September 2018
• Currently positioned as the affordability-led alternative to Pushpay

URL. https://get.tithe.ly/compare

Source 8.1.4: Planning Center | NEW

• Continues as the most widely-used ChMS in the U.S., modular pricing
• Privately held, has not announced significant external funding rounds
• AI positioning is limited compared to Subsplash and Pushpay

8.2 Nonprofit CRM

Source 8.2.1: Virtuous | VERIFIED

• Total funding: $141M across 8 rounds; latest a $100M Series C from Susquehanna Growth Equity, September 19, 2024
• Acquired Momentum (AI-first donor engagement for major gift officers) August 12, 2025
• 2026 product positioning: Virtuous Momentum, Virtuous Insights (predictive intelligence), Virtuous BI
• Released 2026 Nonprofit Fundraising Benchmark Report (April 21, 2026, n=771 mid/large nonprofits)

URLs.

• https://pitchbook.com/profiles/company/103012-21
• https://virtuous.org/blog/virtuous-acquires-momentum/

Source 8.2.2: Bloomerang | NEW

• Acquired Qgiv on January 10, 2024
• Reported revenue ~$12M (Prospeo estimate); estimated valuation ~$38.4M; total funding ~$33M across 2 rounds
• 24,000+ nonprofit customers; ~353 employees
• Has integration with Momentum (now part of Virtuous), creating a competitive dynamic

URLs.

• https://prospeo.io/c/bloomerang-revenue
• https://www.crunchbase.com/organization/bloomerang

Source 8.2.3: Salesforce Nonprofit Cloud / Agentforce Nonprofit | NEW

• Rebranded Nonprofit Cloud to Agentforce Nonprofit in 2025
• Pre-built AI agents (December 2025): Prospect Research Agent (GA), Participant Management Agent (GA), Volunteer Capacity & Coverage Agent (Beta, GA early 2026), Donor Support Agent (Beta spring 2026, GA summer 2026)
• Power of Us Program: 10 free Agentforce Nonprofit licenses for eligible nonprofits

URL. https://www.salesforce.com/news/stories/agentforce-nonprofit-announcement/

8.3 Faith-tech and values-aligned AI

Source 8.3.1: Gloo | VERIFIED

• Listed on Nasdaq (GLOO) in 2025; previously raised $110M in venture funding (summer 2024); IPO filing sought up to $100M
• Serves 140,000+ faith, ministry, and nonprofit leaders; ecosystem reach claimed at 450,000+ organizations
• December 15, 2025: launched Flourishing AI Christian (FAI-C) Benchmark (807 curated questions across seven flourishing dimensions; leading models average 61/100; Faith dimension lowest at 48/100)
• Pat Gelsinger serves as Executive Chairman and Head of Technology
• Recent acquisitions: XRI Global (November 25, 2025, multilingual/voice AI); Westfall Group (announced December 17, 2025; closed January 8, 2026, donor solutions for faith and mission-driven nonprofits)
• Strategic partnership with Barna (State of the Church initiative)

URLs.

• https://gloo.com/press/releases/gloo-unveils-the-first-benchmark-exposing-how-ai-misses-christian-worldview-and-values
• https://www.christianpost.com/news/gloo-seeks-to-raise-100m-in-ipo-to-advance-human-flourishing.html

---

9. Theological and philosophical grounding

The "human formation precedes technological deployment" thesis is well supported by primary theological literature.

• Vatican, Antiqua et Nova (January 28, 2025). The strongest single magisterial document on AI to date. Key paragraphs as cited in Section 6: ¶67 (deskilling and surveillance), ¶105 (humanity enslaved to its own work), ¶112 (AI as complement, not replacement). https://www.vatican.va/roman_curia/congregations/cfaith/documents/rc_ddf_doc_20250128_antiqua-et-nova_en.html

• James K.A. Smith, You Are What You Love (2016). Brazos Press. The accessible companion to Desiring the Kingdom (2009). Direct ground for the formation-vs-information distinction. The argument that institutions are shaped by their liturgies (their repeated formative practices) maps cleanly onto Movemental's claim that AI deployment without formation produces deformative outcomes.

• Andy Crouch, Culture Making (2008) and The Tech-Wise Family (2017). Crouch is on record (as Praxis Partner for Theology and Culture, e.g., the Harvard CCG Christianity and AI conference) as a leading voice connecting Christian formation to technology decisions.

• Pope Leo XIV. Multiple 2025–2026 statements on AI and human dignity, including the January 24 message for the 60th World Day of Social Communications and the address to cardinals (May 2025) noting AI "poses new challenges for the defense of human dignity, justice, and labor."

• Daniel Copeland, Vice President of Research at Barna (Barna/Gloo, February 2026). "Though the majority of practicing Christians remain the most cautious about embracing AI as a spiritual tool, their views are shifting and remain largely uninformed by their pastor… There's a real opportunity here for pastors to disciple their congregants on how to use this technology in a beneficial way."

---

10. Claims to drop or revise

The following changes are required for citation integrity.

Claim 10.1. "About 91% of church leaders use AI in their work" (attributed to Lifeway Research, April 2026). DROP: does not appear in Lifeway's report.

Replacement. "42% of U.S. Protestant pastors are using AI for ministry (10% as regular users, 32% as experimenters), according to Lifeway Research's September 2025 phone survey of 1,003 senior pastors (released April 2026). Across the broader category of church leaders (pastors plus staff), Barna and Pushpay's State of Church Tech 2026 report finds 60% personally use AI at least monthly, but only 33% report their church is using AI in ministry operations."

Claim 10.2. "9% of churches have a formal AI policy" (attributed to Lifeway). REVISE. The verified figure is 5%, from Barna/Pushpay 2026 (n=1,306 church leaders, fielded November–December 2025). Lifeway's survey did not measure organizational policy adoption.

Claim 10.3. "About 25% of churches have already encountered AI-generated scams or impersonation" (attributed to Lifeway). DROP: cannot be located in either Lifeway document.

Replacement. Verified harm reporting from FBI IC3 ($893M AI-attributed fraud in 2025; 22,364 AI-flagged complaints), the AI Incident Database (Incident 1355 documenting multi-state pastor impersonation), Vatican Dicastery for Communication ("dozens" of Pope Leo XIV deepfake reports per day), and WIRED/ASIS pastoral incident reporting from late 2025.

Claim 10.4. "47–76% of nonprofits have no governance framework for AI" (attributed to Forvis Mazars 2026 or similar). QUALIFIED: both endpoints are real but they come from different surveys measuring different things. The 47% (Virtuous 2026) measures nonprofits with no AI governance policy. The 76% (TechSoup 2025) measures nonprofits without a formal AI strategy. Cite both, with sources, rather than presenting them as a single Forvis Mazars finding.

Claim 10.5. "81% of nonprofits report individual staff AI use without organizational policy." VERIFIED with wording adjustment. The precise Virtuous wording is: "81% of organizations report using AI individually and on an ad hoc basis." Use this wording.

Claim 10.6. "Near-universal student AI use across institutions, near-universal absence of an Association of Theological Schools standard." QUALIFIED.

Replacement phrasing. "While student GenAI use approaches near-universal levels in higher education (Stanford HAI 2026), the Association of Theological Schools has not yet promulgated a binding AI standard. ATS, Atla, and In Trust currently address AI through the F[AI]thfully Co-Creating series and microcredentials, leaving most member schools to set their own policies."

Claim 10.7. FBI public warnings about voice-cloning fraud targeting religious and nonprofit organizations. QUALIFIED.

Replacement phrasing. "The FBI has documented voice cloning as a primary AI-fraud vector across its 2024–2025 public service announcements and 2025 IC3 annual report ($893M in AI-attributed losses, with voice cloning embedded in BEC and senior-impersonation schemes). Religious-sector deepfake incidents are documented separately in WIRED, OSV News, ASIS Security Management, and the AI Incident Database."

Claim 10.8. "Board governance frameworks treating AI oversight as a fiduciary responsibility (Forvis Mazars 2026 or similar)." VERIFIED. Forvis Mazars (February 13, 2026) explicitly frames AI governance as expanding the board's traditional duties of Care, Loyalty, and Obedience. Council of Nonprofits 2025 Board Roles and Responsibilities and Nonprofit Quarterly (October 28, 2025, "How Nonprofits Can Resist the AI Efficiency Trap") are the supporting sector-literature citations.

---

11. Methodological caveats and use notes

Three caveats to apply when citing this corpus.

Sample composition. The strongest empirical results come from samples that overrepresent larger and more digitally mature organizations. Virtuous's 346-organization sample skews toward customers and prospects of a fundraising platform; Barna and Pushpay's 1,306 church leaders came roughly half from each organization's network. Both are valuable but are not random national probability samples. Lifeway's 1,003-pastor sample is the cleanest methodologically (stratified random, weighted phone survey, ±3.3% margin of error) and should be the anchor for any specifically Protestant-pastor claim.

Metric incommensurability. Several of the most striking enterprise figures (the BCG 5%, the McKinsey 6%, the MIT NANDA 95%) measure different things. BCG is measuring "value at scale" across 1,250 firms; McKinsey is measuring EBIT impact >5% across 1,993 firms; MIT NANDA is measuring revenue acceleration from individual pilots across a much smaller sample. They tell a consistent directional story but the specific numbers are not interchangeable. When citing, name the specific metric.

Faith-sector incident prevalence. The deepfake/voice-cloning incident reporting is documented in news sources (WIRED, VICE, Gizmodo, OSV News, ASIS, KnowBe4) and the AI Incident Database, but there is not yet a peer-reviewed or government tally of religious-sector incidents specifically. The qualitative case for harm is strong and growing; a specific national prevalence rate for the church sector is not yet available. Treat any single percentage claim about "what share of churches have encountered an AI scam" as unverified until a primary survey publishes one.

General use principle. Apply this corpus the same way a careful analyst would: cite the primary source, name the specific metric, and acknowledge sample composition. The thesis Movemental is making (that adoption is broad, capability gain is rare, governance is the bottleneck, and formation is the missing input) is supported by the strongest 2025–2026 research available across nonprofit, faith, theological, and enterprise sectors. The few specific statistics that need correction are corrected in Section 10.

---

12. Appendix: source index

Primary sources cited in this corpus, organized alphabetically by author or organization.