Could regulating the creators deliver trustworthy AI?

Professional accountability may be necessary but insufficient: The paper questions whether creating a regulated "AI Architect" profession—similar to licensed engineers or doctors—could ensure trustworthy AI, suggesting that professional responsibility alone cannot solve the trustworthiness problem without addressing deeper questions about ethics and standards.

"Trustworthy AI" lacks a clear definition: The paper identifies a fundamental challenge: there is no consensus on what trustworthy AI actually means. Some define it as legal compliance, while others emphasize ethical standards, creating confusion about whose ethics and which standards should apply.

Current AI deployment lacks transparency and accountability: The paper notes that AI is "often deployed in everyday technologies, devices and services without our knowledge," highlighting a critical gap where users cannot identify a responsible party for AI systems affecting their lives.

Legal compliance and ethical standards may conflict or be inadequate: The paper raises the critical question of whether compliance with laws and ethics standards—even when both are met—is actually sufficient to produce trustworthy AI, suggesting that current regulatory frameworks may have inherent limitations.

Defining whose ethics matters is unresolved: A core challenge identified is determining "whose ethics and which standards should be applied," indicating that trustworthy AI regulation requires resolving fundamental questions about values and priorities before professional licensing can be effective.