AI & Neurodivergence: Who Actually Benefits?
Earlier this year a headline made the rounds claiming that only the neurodivergent will survive the AI takeover. It traced back to a remark from Palantir’s CEO, who said in an interview that there were basically two ways to know you have a future in this economy: vocational training, or being neurodivergent. Around the same time, Palantir launched a Neurodivergent Fellowship whose materials say that neurodivergent people are positioned to play an outsized role in the AI era, crediting pattern recognition, non-linear thinking, and hyperfocus.
I work closely with neurodivergent adults, and I have spent a lot of time thinking about AI — not just its promise, but also its risks and its many different use cases. So when I saw this headline, my reaction was mixed.
Part of me understood the appeal. After decades of difference being framed as deficit, here was a high-profile claim framing it as advantage.
But the more I sat with it, the more complicated it became. I did not want to dismiss it outright. There is truth in the idea that some neurodivergent traits can become strengths in the right setting. But I also did not want to accept the headline on its own terms.
A flattering simplification is still a simplification.
So I want to take the claim seriously and also take it apart a little, because both the celebration and the dismissal miss what the evidence actually supports.
What Is True
There is real research suggesting that some neurodivergent people show strengths that can matter in certain kinds of work. Studies of adults with ADHD have found higher self-reported hyperfocus, creativity, and cognitive flexibility, and have linked knowing and using those strengths to better well-being. Workplace research on autistic adults points to attention to detail, principled reasoning, and technical depth. Some studies of dyslexia find advantages in creativity or visuospatial tasks.
There is also a coherent mechanism for why AI specifically might help. For many adults with ADHD, the hardest part is not the thinking. It is the executive scaffolding around it: breaking a task down, getting started, estimating time, drafting the email that has been sitting in your head for a week. Early research and a lot of lived experience suggest AI tools can act as external scaffolding for exactly those bottlenecks.
That is genuinely promising.
What Is Not
What the evidence does not support is the leap from “some people, in some tasks, under the right conditions” to “neurodivergent people, as a class, will win.”
A careful systematic review of the so-called autism advantage at work found no strong evidence either for or against a general workplace advantage. The ADHD literature is at least as strong on strain as on strength: the same hyperfocus that helps can sit right next to chronic stress, overcompensation, and burnout, and inattention is consistently tied to work problems. The dyslexia findings are task-specific and do not generalize cleanly.
In other words, the honest version is conditional.
Outcomes depend on the person, the task, the team, the tools, and the support. The research is far stronger on barriers than on superpowers. Neurodivergent workers report fear of disclosure, pressure to mask, stigma, and managers who do not know how to help. One UK survey found that a majority of neurodivergent employees feared discrimination from management. A large workforce report found roughly a third had not told their manager or HR at all.
Why the Framing Matters Clinically
Here is the part I care about most.
When you tell people a group has a built-in edge, two things tend to follow.
Managers underinvest in accommodations, because why support a population that is supposedly already winning?
And the celebration quietly narrows to a specific, profitable subset of traits: the marketable hyperfocus, the elite pattern recognition, the unusual tolerance for complexity. The parts that need actual support get left out of the story.
Difference gets folded into a labor-market pitch.
The Palantir fellowship materials are explicit that the program is not a diversity initiative, which tells you the frame is strategic, not about removing barriers.
That is the move I would flag for a client, and it is the same distortion I see in a lot of AI talk generally: sorting a complicated thing into all-good or all-bad.
A neurodivergent profile is not a superpower and it is not a defect.
It is a particular way a system is organized, with real strengths and real costs, and whether it flourishes depends heavily on whether the environment fits.
More Capable, or More Dependent?
The same caution applies to the AI tools themselves.
Using them to break down a task, draft a hard message, or structure a plan can be a reasonable support. For many neurodivergent adults, that kind of external structure can reduce friction and make action more possible.
But there is a difference between using a tool to support your thinking and using it to replace your relationship to your own judgment.
That is the line I care about.
Does the tool leave you more capable after using it, or more dependent on it to begin, decide, regulate, or move forward?
Letting AI replace the work of deciding, reflecting, or self-understanding is where the trouble starts.
There are also real concerns worth holding alongside the promise, around privacy, accuracy, and dependence, and the early research on AI as a support for adults with ADHD is genuinely early.
So will neurodivergent people win the AI era?
It is the wrong question.
The better one is the clinical one: under what conditions does a particular person, with a particular profile, actually do well, and what would it take to build those conditions?
AI might make some strengths easier to express and some bottlenecks easier to work around. It does not erase the need for fit, support, and a realistic account of cost.
That version is less exciting as a headline. It is a lot more useful if you are the person living it.
If you are a neurodivergent adult trying to sort the hype from what actually helps, that is a lot of what I do. I keep a free set of resources on using AI well with ADHD at the AI and ADHD Resource Hub. If you want clarity on your own profile rather than a headline about your category, you can read about that work at assessment.wdtherapy.com.
Stephen Andrew Waller-DeLaRosa, LPC, is the founder of WD Therapy in Georgetown, Texas, offering therapy and clinical clarification for neurodivergent adults across Texas. This post is general education, not clinical advice.
A Note on Sources
This piece draws on public reporting of the CEO interview and the company’s own fellowship materials, plus peer-reviewed research on neurodivergence and work, including a systematic review of the autism workplace advantage, ADHD strengths and workplace strain studies, dyslexia creativity and visuospatial research, workplace masking and disclosure studies, and emerging work on AI as executive-function support for adults with ADHD. Professional guidance from the APA and others on AI in mental health informs the caution. A fuller reference list is available on request.
