This piece is adapted from a conversation with Ifeoma Ike on From Chains to Links, the Black Innovation Alliance podcast. I am a Howard graduate. My son attends North Carolina A&T. I am not neutral on this.

The danger is not the technology

Every few months a new wave of commentary arrives about what AI is going to do to Black workers, Black businesses, Black communities. Most of it treats the technology as weather: something that happens to you.

I would put it plainly: AI is a tool and not a strategy. The risk to Black communities is not that the tool exists. It is disengagement from who builds it, who owns it, and who sets the conditions under which it gets deployed.

If we don't get Black folk involved in setting the conditions to engage the data and the algorithms, don't fuss when you can't show up. It's like the old-school EPMD song, "It's like lotto. You got to be in it to win it."

Follow the money, earlier than you think

Here is the part most founders never track. The CHIPS and Science Act moved roughly $53 billion, with about $13 billion routed through the National Science Foundation and the Department of Energy.

By the time that money appears as a solicitation, the important decisions are already made. The research priorities are set. The eligibility language is written. The consortium partners have been talking for a year. Applying at that stage means competing on someone else's terms, defined by people who were in rooms you were not in.

Being at the table where programs get shaped is a different activity from applying for grants, and it is the one that gets neglected. It looks like advisory boards, standards bodies, working groups, and the unglamorous committee work where scope gets decided. I have spent a lot of my career in those rooms, and the pattern is consistent: the institutions that show up early do not have to fight as hard later.

The number that frames everything

A 2022 Forbes investigation found that North Carolina A&T had been underfunded by $3.8 billion between 1987 and 2020, measured against its land-grant counterpart. Over the same period it generated a comparable $2.3 billion in regional economic impact.

Sit with the ratio. Comparable economic output, a $3.8 billion shortfall in the operating base that produced it. That is not a performance gap. It is a capitalization gap, and it compounds, because the missing money is exactly the money that would have built the research infrastructure, the endowment, and the industry relationships that attract the next round.

No single institution closes a hole that size on its own. A coordinated regional engine might. A&T, Winston-Salem State, Johnson C. Smith, and North Carolina Central sit close enough to operate as one innovation economy rather than four under-resourced ones, and together they can pursue work none would win alone. That is an organizing problem, not a funding problem, and organizing problems are solvable. I work on the same design question with universities and HBCU innovation offices.

Ask why twelve to fifteen times

Toyota's Five Whys, formalized by Taiichi Ohno as part of the Toyota Production System, is the version everyone learns. In my experience five is nowhere near enough outside a manufacturing line, where the causal chain is short and physical.

I push founders to ask why twelve to fifteen times. It feels absurd around the seventh, which is roughly where the retelling runs out. That is roughly where it starts working. What a founder brings in as "the problem" is almost always a symptom that has been described so many times it hardened into a fact, and the first several whys just peel off the retelling. The root cause tends to sit somewhere nobody wanted to look, usually in an ownership question or an incentive nobody controls.

Related, and for the same reason: teach business models, not business plans. A plan assumes the future is knowable and rewards the person who writes the most confident document. COVID settled that argument. Institutions that had built the muscle to iterate a model adapted. Institutions with excellent five-year plans did not. If you are choosing what to teach student founders, teach the iteration.

The audacity of imagination

My grandparents migrated from Seneca, South Carolina. They had third and sixth grade educations between them. What they gave the next generation was not capital, because they had none, but permission to imagine at a scale their circumstances did not justify.

Under-invested communities are rarely short on talent or on work ethic. What gets rationed is permission, and the infrastructure that makes ambition survivable. It is the difference between a moon-shot programme and a good idea nobody funds: not the idea, but whether a system exists to carry it.

Own the tech, do not rent it

The consumption trap is one I watch play out constantly. It is entirely possible to be an enthusiastic user of AI tools and hold no position whatsoever in how they work or who profits.

The alternative already exists in practice. Latimer AI was built specifically because the mainstream models carried gaps and distortions in how they handled Black history and context. Students at Howard are building tools rather than waiting for tools to be built for them. That is ownership, and it scales differently from adoption.

It is worth remembering that the ecosystems doing this well produce serious companies. Capella Space, a satellite radar imaging company, came out of the mission-driven entrepreneurship programs I helped run at the Common Mission Project. The programme structure is what made that outcome reachable.

What innovation is actually for

Innovation, for me, is not just about the next best thing for customers. Innovation, for me, is a cultural innovation.

I think better products are the easy part. The harder and more durable work is changing who gets to build, who gets to own, and who is in the room when the conditions are set. That is the right to innovate, and it is not granted. It is taken by showing up early and often, in rooms that look boring from the outside.

If you are building that capacity inside an institution, I have written about how campus programs reach off campus.