David Yi, Audrey Tan & Chuck Boyce — Disarming AI Without Abandoning Innovation | CYOG Special Edition
AI does not arrive with a moral compass of its own. It arrives with capability. The direction comes from us.
When a system can write, decide, predict, persuade, and scale almost anything, the question is no longer only what the technology can do. The harder question is what we are willing to authorize it to do—and what kind of people we become when we hand over the judgment.
That is the tension at the center of this special Create Your Own Gravity conversation with David Yi, Audrey Tan, and Chuck Boyce.
Who was in the room
Joe Moore and Katrena Drake hosted the conversation, bringing together three guests whose perspectives meet at the intersection of technology, education, capital, faith, and human responsibility.
David Yi joined from the perspective of a founder, investor, educator, husband, and father. Audrey Tan brought the view of a founder and investor working with wealth, stewardship, financial wellbeing, and purpose. Chuck Boyce brought more than four decades in the computer and software industry, along with a practical concern for ethical, safe, and risk-aware implementation.
The conversation was prompted by a call to disarm—not only weapons, but the habits, incentives, and forms of language that turn human beings into resources, data points, markets, or targets.
The question beneath the conversation
The deeper question is not whether AI is good or evil. It is whether human beings are willing to remain responsible for what our tools amplify.
The group moved from the boardroom to the classroom, from investment portfolios to children growing up inside an automated world. The unresolved tension was not a flaw in the conversation. It was the point.
Capability is not authorization
A new system can create efficiencies, automate decisions, and open access to information. But every new capability also creates a new decision about boundaries, oversight, and consequence.
Chuck kept returning to the line between what AI can do and what a person or organization should allow it to do. That distinction is easy to lose when the technology moves faster than reflection.
Featured quote 1 — The ability to do something is not permission
“AI is capable of doing many, many, many things, but just because it can, should it be authorized to do it?”
Ethical AI begins before deployment. It begins when someone is still willing to ask whether the result deserves to exist, who carries the consequence, and where the human control belongs.
AI amplifies the person behind the system
David described AI as an amplifier. It can extend creativity, learning, and access to knowledge. It can also extend greed, impatience, violence, and the desire to control.
The technology did not create every problem now associated with it. It gives older human patterns more reach. A tool can be used to build a wider circle of care or a more efficient system of extraction. The machine does not make the moral choice disappear. It makes the choice more consequential.
Featured quote 2 — Responsibility stays human
“AI isn’t evil. But people make AI evil.”
This is why the conversation resisted both easy optimism and easy fear. AI can help people inquire, create, and learn. It can also make harmful decisions faster and hide the human chain of responsibility behind layers of automation.
Disarming can begin with economic choices
Audrey pushed the idea of disarmament into investment and stewardship. The decision is not only what a person says they value. It is also what their money, systems, and attention make easier to build.
Capital carries intention. So does code. So does the decision to optimize for growth at any cost, or to slow down long enough to see who bears the cost of that growth.
Featured quote 3 — Let the portfolio tell the truth
“When the human equates to data and data becomes something that can be commoditized, or then the person becomes commoditized as a consequence of that.”
That line turns ethics into a practice rather than a posture. It asks investors, founders, and builders to look at what their decisions are doing to the sovereignty of the human person and the data attached to that person.
Data is not the same thing as a human being
A person can be represented by data without becoming reducible to data.
The distinction matters for families, children, organizations, and communities. The information people place into a model can be copied, inferred, recombined, and used in ways they did not anticipate. Guarding data is therefore part of guarding the person.
The question is not only whether a system is secure. It is whether the people using it understand what they are surrendering, what they are authorizing, and what they can still take back.
The one-person unicorn may be the wrong ambition
The promise of the one-person company is alluring: one founder, one idea, extraordinary leverage, no friction.
But friction is not always waste. Other people bring correction, memory, humor, moral resistance, and the ability to see what the builder has stopped seeing. A system designed to eliminate every human dependency may also eliminate the relationships that keep ambition honest.
Education is where the future becomes visible
The promise and danger of AI became concrete when the guests discussed education.
AI can help people search a deep body of knowledge and become better thinkers and inquirers. It can also make it easy to skip the struggle through which understanding is formed. If students use a system to produce every answer, they may receive the product of thinking without developing the capacity to think.
That concern extends beyond school. Children are learning what effort feels like, what risk feels like, and whether the physical world still deserves their attention. The question is not simply whether young people know how to use a device. It is whether they are still being invited to build, test, fail, repair, care, and participate in real life.
What the conversation makes possible
This conversation does not end with a policy checklist. It leaves a more demanding invitation: slow down long enough to notice what is being amplified.
Are we using AI to extend human judgment, or to avoid the responsibility of having judgment? Are we making systems more efficient while making people less capable? Are we building tools that widen human dignity, or systems that make exploitation easier to scale?
The market will reward speed. Capital will reward growth. Platforms will reward dependence. The work of ethical AI is to decide where the human being remains non-negotiable.
Frequently asked questions about ethical AI
What is ethical AI?
Ethical AI is the practice of building and using artificial intelligence with human judgment, dignity, safety, accountability, and real-world consequences in view.
Why does human judgment matter in AI?
AI can produce an answer or recommendation, but it cannot carry moral responsibility in the human sense. People must decide what should be authorized, trusted, limited, or stopped.
Can AI be used for good?
Yes. The guests described AI as a tool that can extend learning, inquiry, creativity, and access to knowledge. Its value depends on the intent, safeguards, and human relationships surrounding it.
What should leaders refuse to normalize?
Leaders should resist blind speed, total dependence on automation, the commoditization of personal data, and growth that ignores the people, communities, or environment carrying the cost.
Enter the room
These questions become different when they meet real people: a founder, an investor, an educator, a parent, a builder, a child, a community.
Hear the full Create Your Own Gravity conversation and stay with the questions that do not resolve on command. Ethical AI is not only about the systems we build. It is about the world our decisions make easier to inhabit.
