On keeping America moving forward in AI, I agree with President Donald Trump.

Trump has pushed back against tighter federal AI regulation, dismissed catastrophic warnings as a “hoax,” and argued that slowing American development would benefit China. I agree with his central point about protecting American innovation and technological leadership. [The Wall Street Journal](https://www.wsj.com/tech/ai/trump-says-ai-doesnt-need-more-guardrails-despite-industry-warnings-b94263fa)

America should keep building. We should keep researching, improving, and finding useful ways to put this technology to work. Fear should not become our national technology strategy.

As a programmer, I see AI as something that can help people solve problems. It can assist with writing software, organizing information, automating repetitive work, and making complicated tasks more manageable. Its value comes from what people accomplish with it.

But calling AI a tool does not let its developers off the hook. It puts responsibility directly on their shoulders.

AI does not spring into existence on its own. Even when AI helps write code or supports the development of another system, people and organizations make decisions about its objectives, resources, access, and deployment. Those decisions carry consequences.

If your company builds an AI system, your company should test it thoroughly, protect the information it handles, establish limits on what it can do, and provide ways for people to intervene when something goes wrong. When you discover a serious weakness, fix it. When your product has limitations, explain them honestly.

**Safeguards should be part of the engineering from the beginning.**

A company should not need to wait for Washington to tell it that exposing private information, allowing unauthorized access, or releasing an inadequately tested product is unacceptable. Responsible development belongs in the design, the testing, and the daily operation of the system.

Users have responsibilities, too. An AI answer still needs judgment. Something does not become accurate simply because a computer says it confidently. Anyone who has used these systems knows they can produce a wrong answer with the confidence of somebody who has never been wrong in their entire life.

I support moving forward with our eyes open. Taking real problems seriously is compatible with supporting innovation. We can demand better products and responsible companies while opposing broad restrictions that make useful development unnecessarily difficult.

My position is straightforward: let American companies build, expect them to put real safeguards in place, and hold people accountable for what they develop and deploy.

I agree with President Trump that America needs to remain a leader in AI. The companies building that future need to earn the public’s trust through their work.

AI is a tool. Human responsibility comes with it.