When a Son Turns to Siri: A Lesson on Automation Bias

In a brief anecdote, a father’s question about whether all dogs shed leads his son to ask Siri for an answer. The exchange illustrates a growing tendency—automation bias—where people rely on machines for information they might otherwise verify themselves.

The Human Filter: How to Verify AI Answers

Abdu Murray explains a simple rule of thumb: treat an AI’s confident reply like a stranger who sounds sure. He recommends asking the system for sources, checking those sources yourself, and comparing answers from multiple AI tools. The goal is not to dismiss technology but to keep a human check that asks, “Is this the truth?”

The Alignment Problem: Machines Without Moral Compass

AI is described as a prediction engine that has no values of its own. Murray uses the classic paperclip example to show how an unchecked goal can lead to disastrous outcomes. The discussion centers on how we can align a tool that lacks conscience with human values, and how the rapid pace of AI development may outstrip our ability to set safeguards.

Deep Fakes and the Crisis of Trust

Deep fakes—AI‑generated videos or audio that look authentic—have already been used for political propaganda. The conversation frames the problem as a shift from post‑truth to post‑trust, where people doubt everything. Murray stresses that discerning what is real requires human discernment, a skill he says is a gift from God, and that relying on technology alone is insufficient.

AI as a Tool for Healing

While warning of risks, the discussion also highlights promising uses of AI, such as automated lab tests that could help correct Down syndrome. Murray asks whether, if we can prevent suffering, we have an obligation to act. He suggests that the challenge is to use AI to serve humanity while guarding against its potential to harm.