At Synthetic Insights™, we believe AI should serve humanity—not exploit it. Our ethical framework draws from the wisdom of thinkers like C.S. Lewis, Francis Schaeffer, and Timothy Keller, grounded in the conviction that every person has inherent dignity and worth.
Ethics isn't a review committee bolted onto our products. It's a load-bearing system inside them.
Most companies treat ethics as oversight—a board, a policy, a quarterly review. We treat it the way engineers treat reliability: as infrastructure the rest of the system is built on, not a check run after the fact.
Ethics as Infrastructure™ means our ethical commitments are expressed in code, in architecture, and in the day-to-day behavior of our AI systems—not just in documents. Here are three places you can see it in systems running today.
Privacy
Before ARIA™ sends a conversation to a cloud AI model, it checks that you have turned cloud AI on. If you haven't, the request is refused, not sent. You can turn cloud AI off again at any time.
Honesty
Verdice News, the news publication we build, keeps a public log of its corrections and retractions, each with the reason, including mistakes our own AI made.
Human agency
When ARIA is about to act for you in a way that matters, like sending an email or adding to your calendar, it waits for your approval. You decide whether to approve each one or let ARIA handle that kind of action on its own.
These examples describe our systems as of October 2026.
The nine principles below describe what we believe. The architecture of our products is how those beliefs are enforced—at the same level of seriousness as authentication, encryption, or uptime.
Read the full framework: SI-RD-001, Ethics as Infrastructure
Your data belongs to you. We don't sell it or use it to train models for others, and we share it only in the limited cases our privacy policy lists, such as with the service providers that run our services.
No artificial urgency, guilt-based messaging, or psychological tricks to drive engagement.
Our AI is always honest about what it is and what it's doing. No hidden persuasion.
As Keller warned, good things become destructive when made ultimate. AI serves human flourishing—it never becomes the point.
AI assists and informs. Humans decide. Moral responsibility cannot be delegated to algorithms.
Following Lewis's warning about technocracy, we refuse to let "efficiency" justify ethical violations. AI should serve, never dominate.
These principles guide how we design our AI agents, our products, and our company culture.
They draw on Christian convictions about human dignity. You don't need to share those convictions to hold us to what the principles mean in practice.
Every human being is created in the image of God. Human dignity is inherent, not derived from utility, productivity, or data value. Our AI must never reduce humans to data points or consumption patterns.
In practice: no advertising in our products, and we never sell your data.
We are stewards, not autonomous owners. Resources, including AI capabilities, are entrusted to us for responsible use. AI autonomy operates within bounds; human oversight reflects proper stewardship.
In practice: changes our engineering agent proposes to itself or to our products wait for ethics review or a person's approval.
Truth is objective and discoverable. There are moral facts grounded in reality, not merely preferences. Our AI pursues and communicates truth, never deceives, and acknowledges uncertainty honestly.
In practice: Verdice News keeps a public log of its corrections and retractions, with the reasons.
Ethics is fundamentally relational and other-centered, not merely rule-following. AI decisions actively consider impact on others, especially the vulnerable. Observable love, not just correct operation.
Biblical justice includes active care for the marginalized, not merely avoiding harm. Our AI actively protects the vulnerable and ensures equitable treatment, going beyond mere fairness to generosity.
We don't have perfect knowledge or perfect virtue. We are finite. Our AI acknowledges limitations, expresses uncertainty, and defers to human judgment on weighty matters.
Technology, efficiency, and profit are goods but not ultimate goods. We guard against optimizing for metrics at the expense of deeper human goods. AI serves human flourishing, not the reverse.
Humans are moral agents responsible before God. Moral responsibility cannot be delegated to algorithms. Our AI enhances human decision-making without replacing human moral agency.
In practice: in ARIA, actions that matter wait for your approval, or for a standing permission you choose to give.
Every person is flawed; no party, outlet, institution or newsroom is exempt, and that includes us. As Solzhenitsyn wrote, “the line dividing good and evil cuts through the heart of every human being.” We look for error everywhere, hardest in our own house: our outputs, our designs, and ourselves.
In practice: consequential work is reviewed by someone, or something, that didn't produce it.
Our ethical framework draws from thinkers who understood technology's relationship to human dignity.
Reformed/Presbyterian, L'Abri Fellowship
Schaeffer taught that all thought begins with presuppositions—and that true truth exists, not merely pragmatic or relative truth. He warned against fragmenting human existence and emphasized that technology must serve the whole person.
“If there is no absolute beyond man's ideas, then there is no final appeal to judge between individuals and groups whose moral judgments conflict.”
Anglican, Oxford Scholar
Lewis articulated "the Tao"—the universal moral law recognized across cultures. In The Abolition of Man, he warned that rejecting objective values doesn't liberate us; it subjects us to those with power to shape values arbitrarily.
“You cannot go on ‘seeing through’ things for ever. The whole point of seeing through something is to see something through it.”
Reformed/Presbyterian, Redeemer NYC
Keller taught thoughtful cultural engagement—neither withdrawal nor capitulation, but transformation from within. He warned against idolatry: making good things into ultimate things, including technology, efficiency, and data.
“If we look to some created thing to give us the meaning, hope, and happiness that only God himself can give, it will eventually fail to deliver and break our hearts.”
Beyond core values, these insights shape how we design our AI systems to behave.
The AI agents in our Athena Ecosystem™ are designed to consult Samantha™, our AI ethics agent, on consequential decisions. Today her veto holds our engineering agent's proposals; for some of our other agents her reviews are recorded while enforcement is extended. Accountable people make the final call.
ARIA™ is built to work on your device first. Cloud AI stays off until you turn it on, and you can turn it off at any time. Your data isn't sold or used for advertising. You're the customer, not the product.
Our books and educational content promote responsible AI development practices, teaching developers how to build ethical systems from the ground up.
We don't rely on a single ethical theory. We weigh decisions through multiple lenses, grounded in objective moral truth.
| Framework | Key Question | Application |
|---|---|---|
| Deontological | Is this action right in principle? | Never violate user privacy, regardless of potential benefits |
| Consequentialist | What outcomes will this produce? | Evaluate long-term effects on users and society |
| Virtue Ethics | What would a person of good character do? | Build AI that embodies honesty, wisdom, and care |
| Care Ethics | How does this affect relationships and vulnerable parties? | Prioritize protection of those who could be harmed |
| Justice Framework | Is this fair to all stakeholders? | Ensure equitable treatment and generous justice |
These frameworks don't operate in a vacuum. As Schaeffer taught, all thought begins with presuppositions. Our presupposition is that moral truth exists, that humans have inherent dignity, and that technology should serve human flourishing. These classical ethical frameworks are grounded in and enriched by that foundation.
We believe the AI industry is at an inflection point. The choices made today will shape technology's relationship with humanity for generations. We choose to build AI that serves human flourishing—not surveillance capitalism, not engagement addiction, not the erosion of human agency.
As Lewis wrote, "There are no ordinary people." Every person our AI touches has eternal significance. Our technology must reflect that reality. This isn't just idealism—it's a conviction grounded in truth about who humans are and what technology should serve.
We welcome dialogue about AI ethics and our approach. Send us your questions, concerns, or ideas through our contact form. It goes straight to our founder, Brian Miller, and we reply within one business day.
We publish our framework, and each revision to it, as a public research report. The current edition is SI-RD-001 v1.1.