Three Bullet Points for a Trillion-Dollar Problem Subline

The White House released its National Policy Framework for AI on March 20th.

Four pages. Seven sections. Legislative recommendations for Congress covering child safety, intellectual property, free speech, infrastructure, innovation and federal preemption of state AI laws.

The industry response was immediate. The U.S. Chamber praised it. CTA praised it. Law firms published explainers within days. Most of the conversation centered on federal preemption of the state-by-state patchwork of AI regulation. That’s a real issue and the framework addresses it directly.

Almost nobody talked about Section VI.

⸻

Section VI is titled “Educating Americans and Developing an AI-Ready Workforce.”

It contains three bullet points.

The recommendations: use non-regulatory methods to incorporate AI training into existing education programs. Expand federal study of task-level workforce realignment. Bolster land-grant institutions.

No new programs. No new funding. No employer guidance. No governance frameworks. No mention of organizational readiness.

The framework gets a lot right on innovation and regulatory clarity. But on the question of how companies actually prepare their people for AI, it’s thin. And that’s the part most organizations need the most help with.

⸻

I’m not making a political argument here.

This isn’t about what the framework should have said.

It’s about what companies need regardless of what any framework says.

MIT’s research has mapped the exposure. AI can perform tasks tied to roughly 12% of the U.S. labor market right now. McKinsey puts the theoretical automation potential higher. Brookings has documented the limits of retraining programs when workers bear the time and opportunity costs themselves.

Those numbers are worth understanding, not because they predict mass displacement, but because they clarify where the work is. AI doesn’t arrive as a wave that hits everyone equally. It arrives task by task. Some roles shift. Some roles grow. Some roles become more valuable because the routine parts get handled and the judgment parts become the whole job.

Subscribe now

The companies getting this right aren’t treating it as a threat. They’re treating it as a design problem.

⸻

Here’s what the framework doesn’t address that companies need to build themselves.

Employer-level AI governance. Acceptable use policies. AI councils or cross-functional oversight. Risk mapping for AI use cases. These aren’t optional for any organization that’s serious about adoption. They’re the structure that makes adoption safe and sustainable.

Training architecture that actually works. Not a link to an online course. Role-based training. Segmented fluency paths. Integration with career development. The structural work that turns “we offer AI training” into real capability people can use in their actual jobs.

Organizational readiness. Change management. Honest conversations about what AI does and doesn’t change. The cultural work that determines whether people engage with AI adoption or quietly resist it.

The term “responsible AI” doesn’t appear in the framework. For companies doing this work, it’s the foundation.

⸻

The Fortune 500 has moved fast. JPMorgan is rolling AI tools to 140,000 employees. Walmart invested a billion dollars in skills-first training. Amazon committed $1.2 billion to upskilling.

But the story that matters more is what’s happening at companies without those budgets.

Mid-market firms face the same pressure to adopt AI. They don’t have dedicated AI departments. They don’t have training budgets in the hundreds of millions. What they often do have is something the large enterprises are trying to build from scratch: proximity. Smaller teams. Closer relationships between leadership and the people doing the work. The ability to move with intention rather than bureaucracy.

That’s not a disadvantage. That’s a design advantage if you use it.

The companies I watch most closely aren’t the ones spending the most. They’re the ones building governance and fluency from the inside out, with their people in the room.

⸻

What makes the framework’s gap more visible is what already exists in parallel.

The Department of Labor released its own AI Literacy Framework five weeks before the White House document. It defines content areas and delivery principles. It launched a free AI literacy course. The July 2025 AI Action Plan created a workforce research hub at DOL and directed rapid retraining funding for displaced workers.

At the state level, California’s community college system committed $22 million to AI infrastructure with partnerships across Nvidia, Google, OpenAI and Microsoft. Tennessee became the first state to use MIT’s exposure data in an official AI workforce action plan. Virginia partnered with Google to upskill 10,000 residents.

In Congress, the bipartisan AI Workforce PREPARE Act would require employer reporting when AI is a substantial factor in mass layoffs. The AI Workforce Training Act proposes a 30% tax credit for qualified AI training expenses.

The ecosystem is more developed than the framework suggests. The framework just doesn’t connect to it yet.

⸻

If you’re running AI adoption at your company, here’s the diagnostic.

Do you have a responsible AI policy? Not a vendor’s template. Yours. One your people helped shape.

Do you have a training architecture that segments by role and builds over time? Or do you have a link to a course nobody finishes?

Do you have a governance body with cross-functional representation that reviews AI use cases before deployment?

Do you know which tasks in your organization are most exposed to AI? Not job titles. Tasks.

Do your people trust the process enough to raise their hand when something isn’t working?

Those questions matter more than anything in the federal framework. And the answers come from inside your organization, not from Washington.

⸻

The framework signals direction. Innovation. Infrastructure. Regulatory clarity.

It doesn’t signal workforce readiness. Not yet.

That means the companies doing this well are building the capability themselves. Training their people. Standing up governance. Having honest conversations about what’s changing and what isn’t.

That’s not a gap to worry about. That’s an advantage to build.

The organizations that figure this out won’t be the biggest. They’ll be the ones that brought their people into the process early and built something together.

That work has already started. It just isn’t coming from a framework.