Six Assumptions

Assumptions
Inform but not replace human agency
Develop rather than diminish human cognitive capacity
Respect and advance human embodiment
Serve personal relationships rather than replace them
Restore trust in human institutions
Benefit the global majority rather than enrich and entrench a narrow minority

We begin with these six assumptions. We are well aware that each requires a measure of faith — that is, each is based on a certain amount of evidence that warrants our belief, though none can be proven beyond a reasonable doubt.

The enduring image
A pattern of deceptions
Very good and also very distorted
As consequential as the Internet — or electricity — or agriculture
Asymmetrical risk—even without a singularity
The fantasy of the superhuman
1
The Enduring Image

Human beings uniquely bear the image of God and will continue to do so throughout history. Our ability to discover patterns in the universe, develop science, and create technology reflects this divine vocation. These advancements are meant to glorify God, serve humanity, and contribute to the flourishing of creation itself.

2
A Pattern of Deceptions

Throughout history, humanity has fallen into patterns of deception that distort God’s image, corrupt wisdom, and misuse science. We have sought independence from God, believing we can secure good without relationship, often replacing love with technology and wealth with Mammon. These false pursuits dishonor God, harm people, and exploit creation.

3
Very Good, Yet Deeply Distorted

AI, like all technology, is part of the “very good” world humanity is called to cultivate. It extends human intelligence, incorporating vast cultural knowledge, much like writing once did. However, it also inherits human biases and is subject to deception, shaping both its outputs and its users in ways that may lead to corruption.

4
A Technological Shift as Profound as the Internet

AI’s long-term impact is uncertain, but it may be as transformative as the Internet—or even as foundational as electricity or agriculture. Each of these technologies expanded human potential but also caused unforeseen harm when guided by pride and greed. AI will likely follow this pattern, offering both great promise and significant risk.

5
Asymmetrical Risks, Even Without a Singularity

There is no evidence that AI will replace humanity’s unique role, though it will surpass human ability in specific tasks. While fears of AI “replacing” us are exaggerated, concerns about its misuse are valid. Like nuclear fission, AI has immense potential for both progress and destruction, depending on how it is developed and deployed.

6
The Fantasy of the Superhuman

AI may or may not rival history’s most transformative technologies, but it carries unique risks due to its resemblance to human intelligence. It can mimic traits like personality and purposefulness, though likely without true understanding. Even if AI is less powerful than imagined, the mere fantasy of “superhuman” creation can distort relationships, economies, and societies, amplifying the dangers of pride, magic, and Mammon.

Six Redemptive Directions

Redemptive
Inform but not replace human agency
Develop rather than diminish human cognitive capacity
Respect and advance human embodiment
Serve personal relationships rather than replace them
Restore trust in human institutions
Benefit the global majority rather than enrich and entrench a narrow minority

With these assumptions, we offer guidance for building AI ventures that go beyond ethics to actively repair harm from past technologies.

This guidance focuses on ventures rather than policy. While government and major infrastructure providers shape AI’s trajectory, real change often comes “from below,” where entrepreneurs drive innovation through specific applications.

Like the Internet, electricity, and agriculture, AI is a powerful tool. Redemptive entrepreneurs can lead by showing that AI is best used to counter pride, magic, and Mammon while uplifting human dignity and flourishing.

Inform but not replace human agency.
Develop rather than diminish human cognitive capacity.
Respect and advance human embodiment.
Serve personal relationships rather than replace them.
Restore trust in human institutions.
Benefit the global majority rather than enrich and entrench a narrow minority.
1
Redemptive AI will empower human agency, not replace it.

AI’s predictive capabilities can inform human decision-making but must not be used to limit free and wise choices. In many areas of life—such as policing, credit decisions, and resource management—AI should support, not override, human responsibility. Redemptive AI will restore agency rather than suppress it, ensuring people remain conscious stewards of human dignity and the world entrusted to us

2
Redemptive AI will enhance human learning, not diminish it.

Education is already unequal and often fails to develop human potential. AI could worsen this by offering shortcuts to knowledge or distracting with endless entertainment. Instead, Redemptive AI will strengthen cognitive development by scaffolding learning, supporting skill-building, and promoting wisdom—helping people grow into informed, creative contributors to society.

3
Redemptive AI will honor the human body, not undermine it.

Modern technology has encouraged disembodiment, prioritizing screen-based interactions over real-world engagement. Redemptive AI will instead respect human physicality by integrating more natural interfaces, supporting movement, and enhancing accessibility. It will expand, rather than erode, people’s ability to fully participate in the world around them.

4
Redemptive AI will support relationships, not replace them.

AI can simulate human connection, but artificial companionship is no substitute for real relationships. Redemptive AI will never present itself as a person, nor relieve users of the effort required to truly care for others. Instead, it will facilitate healthier human interactions, ensuring technology strengthens, rather than supplants, personal bonds.

5
Redemptive AI will restore trust in institutions through transparency.

Today’s systems often conceal their operations while exposing individuals to surveillance. AI could worsen this imbalance, further concentrating power among large corporations and governments. Redemptive AI will reverse this trend by making institutions more transparent while protecting individual privacy—ensuring people are known by choice, not by constant data extraction.

6
Redemptive AI will serve the global majority, not just the privileged few.

AI development is capital- and data-intensive, favoring wealthier nations and corporations. Without intentional change, its greatest benefits will flow to those who already hold power. Redemptive AI will focus on unlocking value for the underserved, ensuring innovation uplifts all people rather than deepening inequality.

Six Key Assumptions

1

Enduring Image

Humans uniquely bear God's image

2

Pattern of Deceptions

Tech enables independence without relationships

3

Very Good & Distorted

AI is good but prone to corruption

4

Historical Scale

AI rivals internet, electricity, agriculture

5

Asymmetric Risk

Real risks, but no singularity coming

6

Fantasy vs Reality

AI mimics humans, creating unique risks

Six Redemptive Directions

1

Human Agency

Redemptive AI will inform but not replace human agency.

2

Learning

Redemptive AI will develop rather than diminish human cognitive capacity.

3

Embodiment

Redemptive AI will respect and advance human embodiment.

4

Relationships

Redemptive AI will serve personal relationships rather than replace them.

5

Trust & Privacy

Redemptive AI will restore trust in human institutions by protecting privacy and advancing transparency.

6

Global Equity

Redemptive AI will benefit the global majority rather than enrich and retrench a narrow minority.

About Redemptive AI

The Redemptive Thesis for Artificial Intelligence is a working document to spark the imagination of redemptive entrepreneurs and builders in artificial intelligence. It was authored by Praxis Partner Andy Crouch, in collaboration with Praxis Venture Partner Mark Sears and Praxis Co-Founder & CEO Dave Blanchard, as well as attendees at the Redemptive AI Forum.

See the full version here.

About Praxis

Praxis is a venture-building ecosystem with a redemptive imagination, supporting founders, funders, and innovators motivated by their faith to address the major issues of our time.

More at praxis.co

About Sprout AI Studio

Sprout AI Studio is a venture studio building commercial AI companies that enhance rather than erode human connection and relationships.

More at sproutai.com