When the psychiatrist Leo Kanner first defined autism in 1943, it was estimated that 4 to 5 children per 10,000 were affected by the condition. Today, the Centers for Disease Control puts that number at 1 in 36, or almost one child in every classroom. If any other medical condition—blindness, epilepsy, or paralysis—showed a spike like this, it would trigger a pandemic-level outcry. But after parental panic across the West and a search for a causal villain in vaccines that followed reports of 1 in 166 children in America with the condition in the late 1990s, today the story is very different, despite a supposed quintupling of the number of autistic children in the past 25 years. Now, there is more curious and far less emotionally wrought murmuring as to what autism might be—and even more surprising, perhaps, a growing chorus of people insisting that they, too, should be considered autistic. Now, from experts and parents alike, we get calls for inclusivity as psychiatrists and others praise themselves for having advanced so profoundly in their ability to diagnose the condition and understand it better.

Helped along by characters like Sheldon Cooper on the new century’s most popular comedy program, The Big Bang Theory, and the up-to-date version of Sherlock Holmes as embodied in the wildly popular BBC series Sherlock, being autistic has become something of a badge of honor. You’re socially odd, intellectually superior, and emotionally detached in an edgy and endearing way. For many, especially mothers with narcissistic tendencies who are themselves hungry to possess a narrative of their own exceptionalism, the reframing of autism was too seductive to pass up—a way of casting a child’s concerning behavior and obvious nonconformity not as problems to be overcome but as evidence of special gifts. A devoted parent of an autistic child could thus become the steward of a quirky and special genius. Given the difficulties of raising a child with myriad such difficulties, she could be viewed both as a patient and exceptional person as well as one of the overseers of an elite cultural phenomenon. And psychiatric professionals have made it easy for parents of this particular sort by flinging open the diagnostic gates. Rather than refining and focusing on specific symptoms to make it possible to spot the differences between quirkiness and genuine difference, they have expanded to a breaking point the Diagnostic and Statistical Manual of Mental Disorders (DSM). Once a club no one wanted to join, once a condition that no one would ever have wished on a child, as you would not wish a chronic and incurable condition on any child, autism has become something else entirely: something oddly desirable.

Kanner’s original understanding of autism was fairly rigid and recognized a distinct group with unmistakable qualities: nonverbal children, socially disconnected, cognitively impaired, often with seizures. These were not quirky introverts. These were children who required full-time care and specialized schooling. In the DSM-III of the 1980s, it was called “infantile autism.” The criteria required that it manifest itself before a child was 30 months old, and the condition was recognizable by marked language delays, gross deficits in social interaction, and repetitive behaviors. These were developmental dysfunctions, not misunderstood personalities. And neither clinicians nor parents had a problem naming them as such.

Then came the DSM-III-R in 1987, which introduced the term “pervasive developmental disorder not otherwise specified (PDD-NOS)” and broadened the field significantly. Suddenly, language delay and intellectual disability were no longer central to a diagnosis. Subclinical cases were included. In 1994’s DSM-IV, Asperger’s syndrome was added, which broadened the definition to high-IQ individuals with no language delays but poor social functioning. A child who spoke on schedule but didn’t understand jokes, had poor eye contact, and rigid routines was now also autistic.

The most dramatic change came with DSM-5 in 2013. Subtypes were eliminated. Autism became a spectrum. The criteria were thinned down to two domains: first, social-communication difficulties; second, restrictive and repetitive behaviors. A person needed to meet just six out of 12 traits, spread across these two clusters. Language and cognitive delay? Optional. Even the requirement for early onset was removed. A diagnosis could now be given based on historical symptoms. Questionnaires like the Autism Spectrum Quotient (AQ) are so broad and subjective that a 30-year-old who recalls feeling “socially overwhelmed” in school and not liking itchy clothing can receive the same diagnosis as a nonverbal child requiring lifelong care.

_____________

The diagnostic category has become a black hole, pulling in people with no clinical resemblance, collapsing distinction into sameness. From what I’ve observed, three distinct autism “patients” now account for much of the increased prevalence, none of whom would have qualified under the original criteria.

The first is the temperamentally awkward and quiet child. He is highly conscientious, literal-minded, and has a strong preference for things rather than people. He can spend hours absorbed in intricate play and needs some form of coercion to focus his attention elsewhere. In subsequent forced social settings, he comes across as “weird.” Such a child can be embarrassing for a highly self-conscious parent worried about appearances. What would such a person rather have—a child with an introverted and eccentric personality or a child with “special needs” that might suggest an elevated intelligence and perspicacity?

I keep referring to parents and their conduct here because there is extensive evidence across the internet of just such people writing blogs, making TikToks, and filming YouTube videos chronicling a person’s “autism journey”—both as a parent and as the child himself.

One striking example of how blurred diagnostic categories have become is in the interpretation of “stimming,” a key defining feature of autism. Stimming is an involuntary, neurologically driven motor response to sensory overload: repetitive, unconscious, and difficult to suppress. The thing is, the stimming that’s described on social media by many parents highlighting their own experiences and those of their children is indistinguishable from what any typical two- or three-year-old does: hand-flapping, spinning, lining things up. Such behavior can, to a certain degree, be found in all people. So what’s now presented as evidence of “neurodivergence” is often just a developmental phase that is later reframed through the influence of online checklists and the retroactive placement of unreliable memories.

A common thread in these stories is the initial dismissal by general pediatricians who are extensively trained to distinguish normal developmental variation. After enough doctor-shopping and vague questionnaires, the parent eventually finds someone willing to confirm what she’s already decided: that the child’s aloofness is hidden brilliance. The diagnosis is finally secured by an ideological autism center clinician who is trained not to evaluate critically but to recognize and affirm such “symptoms.”

The second is the undisciplined child—verbally skilled, highly functional, but socially inept. Pediatricians often see nothing wrong with such a kid, because there isn’t a developmental delay. But in the classroom, the behavior becomes disruptive, teachers grow exhausted, and parents give up. Between the second and fourth grade, a diagnosis is sought, not from a clinical need, but from institutional surrender.

The third is the adult Cluster B case. On TikTok, the Cluster Bs proudly recount their autism diagnosis as if it were an Ivy League diploma. To the trained eye, what’s on display is emotional immaturity, narcissistic attention-seeking (“I am the best at being oppressed because I’m special”), and interpersonal manipulation. Their immaturity is revealed in the way they label the rest of society as “phony” and “fake,” casting themselves as the only ones authentic enough to speak the truth, regardless of its impact. It’s the same adolescent script performed since the dawn of time. Only now, the diagnosis confirms the performers’ righteousness. They’re told that they’re wired differently, that their discomfort is insight, and their perspective a rare gift to be bestowed on society. Thus, maturation is stalled.

The performative tendencies of Cluster B patients have them mimicking classic autistic behaviors, such as stimming. But what these patients call stimming is often a form of self-soothing taught in therapy, where repetitive motion and sensory grounding techniques are often successfully deployed to manage emotional dysregulation. And there’s a critical difference. In autism, stimming is a symptom of underlying neurological dysfunction. For others, it is a practiced coping mechanism. The behaviors may look the same on the surface, but they emerge from entirely different clinical realities.

Any attempt to question these realities is swiftly met with the concept of “masking.” If you don’t meet diagnostic criteria, it’s not because you’re not autistic; it’s because you’ve been too good at hiding it. Your true self, oppressed by a society too old-fashioned to see differences, learns all too well how to behave falsely to get along. But the research on masking is methodologically flawed, relying heavily on interpretations of retrospective self-reports from adults diagnosed in their 30s, samples drawn from those already invested in the diagnosis.

It’s circular: If you say you’ve been masking, that becomes proof enough that you were autistic all along. Most people would consider masking no different from the politeness I educate my three small children to show—an ability to put aside their immediate whims to function in a society one day. It’s not evidence of hidden pathology. It’s socialization.

All of this has been made more appealing by the transformation of autism from something considered a dysfunction to something considered a “difference.” This is a deliberate move away from the medical model of disability on the part of what has become a movement—the neurodiversity movement. In the 1990s, the sociologist Judy Singer coined the term “neurodiversity” to frame autism as a natural variation of the human mind. No longer was it to be considered a developmental disorder; no, it was a “neurotype.” Sounds good, and serious, and seems to be the result of sustained longitudinal research, perhaps, but the term “neurodivergence” crumbles under the slightest scrutiny. There is no clinically defined “neurotypical” brain and no bimodal distribution that separates autistic from non-autistic minds.

Groups like the Autistic Self Advocacy Network ran with the unscientific term and lobbied fiercely to widen the diagnosis, especially during the DSM-5 revisions in the early 2010s. They demanded the inclusion of “historical” symptoms, which has enabled adult self-diagnosis. Their mission wasn’t diagnostic precision, but rather diagnostic access.

There are several such activist groups, each playing its role in shifting focus away from those who are genuinely autistic and the needs they have that must be addressed. But when autism is thought of not as an impairment to be treated, but as a unique perspective to be honed, it becomes something else. Under this framework, to describe someone as “low-functioning” is offensive, and the effort to achieve behavioral modification is recast as “ableism.” There is no hierarchy of function to distinguish those whose lives are truly disrupted by the mysterious affliction that affects their physical movements and their ability to connect to others in the world. Instead, they are possessors of diverse minds, and it should therefore follow, we are told, that autists should be offered neurodiversity-affirming interventions and programs tailored to their strengths.

In this vein, another idea that’s gained traction under the neurodiversity banner is the “double empathy problem.” The theory claims that communication difficulties between autistic and non-autistic people aren’t due to a lack of empathy or social skill in the autistic individual, but to a mismatch in communication styles between the two groups. They claim autistic people understand each other just fine; it’s only when interacting with “neurotypicals” that problems arise. What was once viewed as a core deficit in social reciprocity is reinterpreted as mutual misunderstanding. Leave aside that the research in this is weak. The real problem is that it flips the burden of adaptation. “Why should the burden be on the autistic person to make himself understood when the fault might simply lie with the non-autistic person’s failure to get it?” This is not a way to create incentives for parents of autistic kids to take measures, often very difficult and exacting measures, to improve their children’s ability to function in the real world. Which is, in the end, the intended effect.

This was perfectly illustrated in a recent conversation between Jordan B. Peterson and Dr. Simon Baron-Cohen, one of the key architects of the modern autism framework. Peterson pressed Baron-Cohen to define what “severe autism” looks like. Baron-Cohen deflected, saying autism is a spectrum and “always looks different.” This is a pitch for dismantling hierarchy. If you insist that gradations of function are diagnostically relevant, you are supposedly merely demonstrating your lack of understanding on the subject. Baron-Cohen’s answers revealed how ideology can easily overshadow clinical clarity. Vague empathy trumps the precise language we need to create categories that will help us identify correct treatment. The ones most affected by this euphemistic fog are not the high-functioning adults demanding validation, it’s the children who still cannot speak, and whose parents are begging to know: Will they ever be able to live independently?

_____________

This ideological shift away from functioning hasn’t just changed how we diagnose. It has changed how we treat.

In the classical model, when autism was widely accepted as a developmental disorder, treatment was structured, measurable, and goal oriented. It aimed at improving functionality: language acquisition, daily living skills, social adaptation. Applied Behavior Analysis (ABA), a form of managing autism treatment that has been a subject of great controversy, was built around this principle. It broke tasks down into manageable steps and used reinforcement to shape behavior. This was used in conjunction with speech therapy, occupational therapy, and special education—all targeted interventions in which progress could be tracked. The goal wasn’t to eliminate autism, which may not be possible: It was to reduce dysfunction and build independence. And we have collected solid evidence that it works.

Now there is “floor time,” a therapeutic approach that supposedly meets the child where he is. Also called DIR (for Developmental, Individual-differences, and Relationship-based), it is marketed as a humane alternative to the supposedly oppressive structure of ABA. But floor-time DIR has no measurable outcomes, no behavioral goals, and shows little evidence of progress. It relies on vague concepts like “attunement” and “connection,” and therapists are instructed to avoid correction and excessive coaxing. Children are encouraged to “unmask,” to stim freely, with meltdowns and aggression viewed as forms of communication. “Sensory rooms” are prioritized over skill-building and parents are coached to be 24/7 facilitators of co-regulation, abandoning discipline altogether. This approach places enormous emotional burdens on already overwhelmed families, while leaving the most impaired children to flounder, missing key developmental windows where behavior can be learned.

Autists themselves are not the only ones harmed by the consequences of the obsession with inclusivity. In schools, children with severe behavioral challenges are now placed full-time in mainstream classrooms. Paraprofessionals with minimal training act as substitutes for specialized care. In the United States, special-education spending per student with disabilities ranges from $10,500 to $20,100, nearly double the cost for general-education students. Roughly 7.5 million U.S. students receive special-ed services, about 15 percent of the total school-age population. Where districts are lucky enough to have trained special-ed teachers, turnover is high, with burnout cited as a primary reason for leaving.

Not only does this exhaust the staff, but “neurotypical” classmates are forced to tolerate disrupted learning and even bodily harm. Complaints after being spat on or kicked at are met with demands for greater understanding and empathy.

For all the compassion that progressive psychologists, psychiatrists, social scientists, and activists driving this shift are supposedly “modeling” for the rest of us, convincing autists and non-autists alike that society is wrong to expect basic conformity to common codes of conduct isn’t kindness. It’s cruelty. It’s cruel to the truly autistic children who need structured care and specialized environments, not resentful stares from classmates. It’s cruel to the Cluster B patients who require a diagnosis directing them to Dialectical Behavior Therapy, a way of helping those suffering from emotional dysregulation to find a path to living with others and getting their needs properly met as part of a community. It’s cruel to the awkward, odd child who simply needs to be left alone to grow—rather than pathologized. And it’s cruel to the undisciplined child who needs boundaries firmly enforced, not a diagnosis that grants him access to a “sensory room” and even fewer expectations. This is not the way.

We want to hear your thoughts about this article. Click here to send a letter to the editor.