Brad Smith's AI Policy Warning Deserves More Scrutiny Than It's Getting

NewsOnScale Staff

August 3, 2026

Microsoft President Brad Smith made headlines this week by describing the current state of U.S. AI policy as 'regulation without transparent or complete rules' — a pointed critique of an approach that relies heavily on agency guidance, executive orders, and after-the-fact enforcement rather than durable statutory law.

On its face, the observation is accurate. The United States does not have a comprehensive federal AI statute. Oversight is fragmented across the FTC, NIST, sector-specific regulators, and a patchwork of state laws that are multiplying faster than any business can track. The Biden-era executive order on AI established reporting requirements and safety standards, but those can be rescinded with a new administration — as some already have been. The result is a governance environment where companies are genuinely uncertain about what compliance means.

Smith's point lands. The problem is who is making it, and why it matters.

## The Messenger Problem

Microsoft is not a neutral observer of AI governance. The company has invested billions in OpenAI, embedded AI systems across its enterprise software stack, and actively shapes the policy environment it is now criticizing. When Smith argues that ambiguous rules create problems, he is simultaneously correct about the dysfunction and positioned to benefit from a particular resolution to it.

Clear rules, in some scenarios, favor incumbents who can absorb compliance costs and help write the standards. Regulatory ambiguity, in other scenarios, enables rapid deployment without accountability. Smith is criticizing one version of uncertainty — the kind that creates legal risk for large platforms — while his company has historically been less vocal about a different kind of uncertainty: the kind that leaves users, workers, and smaller competitors without meaningful protections.

This isn't a conspiracy. It's just the normal operation of corporate political influence, and it deserves to be named plainly.

## What Transparent Rules Would Actually Require

If we take the call for transparent AI governance seriously — as we should, because it is the right policy goal — then the conversation has to go further than Smith took it.

Transparent rules would mean mandatory disclosure when AI systems are used in consequential decisions: hiring, lending, healthcare triage, content moderation. They would mean audit rights for regulators, not just voluntary safety commitments. They would mean liability frameworks that don't place the entire burden of harm on end users or downstream developers. And they would mean procurement standards that apply to government contracts — where Microsoft is one of the largest vendors.

The Brookings Institution, which appeared twice in this week's policy headlines, has been mapping the fragmentation problem methodically. Their work shows that the U.S. is not simply behind on AI governance — it is operating with a deliberately minimalist federal posture while states and courts fill the gap unevenly. That's a structural choice, not an accident, and it has benefited the largest platforms throughout the cloud and social media eras.

## The Accountability Gap Is the Story

What Smith's comments reveal most clearly is that even the industry's most polished voices now recognize the status quo isn't sustainable. If the company that has done more than almost any other to accelerate AI deployment is saying the regulatory environment is incoherent, that's worth taking seriously — not as a lobbying signal, but as a diagnostic.

The question Congress and the public should be asking isn't just 'should there be clearer rules?' The question is: clear rules that protect whom, enforced by whom, with what penalties for whom?

Those details are where AI governance either becomes meaningful or becomes theater. And the companies most eager to participate in writing those rules have the most to gain from getting that balance wrong — in either direction.

NewsOnScale will continue tracking the gap between AI policy rhetoric and enforceable accountability standards as the legislative calendar develops.