Who Gets to Benefit from AI?

By Burt Brooks, Founder & CEO, PathAble AI

There is a conversation happening right now about artificial intelligence, and it is getting loud. Governments are holding hearings. Influencers are sounding alarms. Tech companies are running ads. Everyone seems to have an opinion, and most of those opinions come pre-loaded with a verdict: AI is either the greatest thing that has ever happened to us or the beginning of the end.

I want to talk about the people who are largely missing from that conversation.

I spent more than twenty years working in disability services before I started PathAble. In that time, I watched the disability community get left out of nearly every major technological shift, not out of malice usually, but out of oversight. People designing systems did not think about us. People debating the future did not include us. And by the time the dust settled, the tools that could have changed lives were either inaccessible, unaffordable, or simply built for someone else.

We are at that crossroads again. And this time, I think we have to speak up before the window closes.

The current backlash against AI is understandable. There are real concerns about surveillance, job displacement, misinformation, the environmental toll of the data centers all of this runs on, and the concentration of power in the hands of a very small number of very wealthy companies. I am not here to dismiss any of that. Hold bad actors accountable. Push back hard. That is important work.

But here is what gets lost when we decide the whole category is suspect: the disability community has been quietly benefiting from AI for years, and almost no one talks about it that way.

Text-to-speech technology is AI. It has given a voice to people who could not speak, opened books to people who could not read standard print, and made communication possible in ways that were unthinkable a generation ago. Autonomous vehicle research is also AI. For someone who cannot drive due to a physical or cognitive disability, the ability to get to work, a doctor's appointment, or just a grocery store independently is not a convenience. It is a civil rights issue. Predictive communication tools are AI as well. These applications learn how a person communicates and help them express themselves faster and more accurately, giving people with complex communication needs a more fluid connection to the world around them.

These are not hypothetical futures. They are happening now. They exist because someone applied this technology thoughtfully, with a specific population in mind.

That is exactly what PathAble is trying to do, building employment support tools for job seekers with disabilities and the staff who work alongside them. We are not alone. ReedAI is helping neurodiverse children develop language skills faster. Eleplan is helping caregivers deliver a better quality of support. There is a growing group of builders who believe AI can serve as a genuine equalizer, particularly around work and community access.

But that work gets harder when the entire conversation about AI collapses into a binary. It is too easy to paint every application with the same broad brush. And when we do, we hand the microphone back to the biggest players, the ones with the loudest PR budgets and the least to lose.

It also helps to be precise about what we are actually talking about. Somewhere along the way we started treating AI like a force of nature instead of what it is: the next step in software. We still have to program it, customize it, monitor it, and work with it, the same as any tool we have ever used. Until something like true general intelligence actually arrives, the job is not all that different than it was before. It is just faster and more flexible, which is what we have always wanted from our software. Naming it plainly takes some of the heat out of the panic, and it lets us judge each application the way we judge any other tool: does it help the person in front of us, or not?

There is one more thing I want to name, because leaving it out would be its own kind of dishonesty: the environmental cost. The data centers that power these tools use real energy and real water, and that lands on real communities. I am not going to pretend PathAble has the answer there. It is not our area of expertise, and I would rather say that plainly than wave it away. What I do believe is simple. AI infrastructure should be built ethically, with the smallest possible footprint on the communities and the environment around it. How we get there will be decided by the experts in that field, not by me. But the rest of us still have a role. We can insist on people over profits. We can let organizations choose developers based on their plans, their approach, and their track record. Nearly every gain in clean energy came from competition over who could do it better and cleaner. Data centers should not be exempt from that same pressure, or from public criticism.

The disability community cannot afford for the conversation to collapse into noise. We have spent too long waiting to be included in progress. I am not willing to let a political tug-of-war over technology determine whether that inclusion happens.

So yes, push back on the bad. Name it. Fight it. And in the same breath, lift up the work that is actually trying to change something.

Because for me, the answer to who benefits is not abstract. It is the job seeker who finally has a way to communicate in an interview. It is the direct support professional who gets an hour back to spend with a person instead of a form. It is the family that stops waiting for a system to catch up to them. Those are the people I am building for, and they cannot afford for us to get this question wrong. So let's answer it together. And let's make sure they are in the room when we do.