The Case Against Toxic Empathy: Why Choosing Who Deserves Compassion Has Become Culture War Currency

A viral video by evolutionary psychologist Gad Saad has reignited a debate that has quietly reshaped how the online intellectual class thinks about compassion, moral obligation, and who deserves help.

By Monexus Staff Writerglobal6-minute read16 May 2026☆ Save ↗ Share ⎙ Print

In a video posted to X on 16 May 2026, Gad Saad offered a provocation that has since circulated well beyond his usual audience of marketing professionals and pop-science followers. Empathy, he argued, is not inherently virtuous. When it hyper-fires — when it becomes indiscriminate, untethered from evolutionary logic, or directed at the wrong targets — it becomes actively harmful. The video drew more than two million views within hours of posting, seeding a debate that has since spilled across podcast feeds, Substack comment sections, and a dozen group chats among people who describe themselves, with varying degrees of irony, as members of the online intellectual class.

The argument is not new. But the timing matters. It arrives in a cultural moment when the vocabulary of empathy has become one of the primary terrains on which political argument is fought. What counts as legitimate compassion, who qualifies as a deserving recipient of public concern, and whether empathy that serves strategic or tribal purposes deserves the name at all — these questions have become load-bearing for how both the intellectual right and its critics define themselves.

What toxic empathy actually means

The concept, as Saad frames it, rests on a distinction that most people intuitively recognize even if they lack the language for it. Ordinary empathy — the capacity to model another person's mental state and feel something in response — is, by most accounts, a functional feature of human social organisation. Evolutionary psychologists have long argued that coalitional thinking, kin selection, and reciprocal altruism provided the phylogenetic substrate from which empathic response eventually emerged. Humans who could read danger in a companion's face, who could anticipate the needs of a sick child, who could calibrate trust based on observed emotional states, survived at higher rates. Empathy is not cultural artefact; it is evolved infrastructure.

Toxic empathy, in Saad's formulation, is what happens when this system malfunctions. The hyper-firing variant occurs when empathic response is triggered by stimuli that do not represent genuine kinship, reciprocal obligation, or genuine harm — when, for instance, social media amplification or activist framing selectively activates moral emotion toward one group while suppressing it toward another. The wrong-target variant occurs when empathy flows toward agents of bad faith — toward those who will deploy received compassion as a resource to be extracted, or who operate on a logic that treats the empathizer as marks rather than as moral equals.

The framing is blunt by design. Saad, a professor of marketing at Concordia University whose academic work focuses on evolutionary psychology and consumer behaviour, has built a following in part because he treats the public intellectual ecosystem as a domain where ordinary social graces do not apply. What he is describing is not a subtle phenomenon. It is the thing that happens when someone gives money to a charity that turns out to be a shell operation, or volunteers compassion for a cause that turns out to be a career vehicle for its founders, or extends moral credit to a political movement that will return the favour by excluding them the moment they become inconvenient.

Why the argument keeps finding new audiences

There are structural reasons this particular formulation keeps breaking through to new audiences. One is algorithmic. Platforms that reward emotional engagement have a systematic bias toward content that activates moral outrage and moral indignation simultaneously — that makes the reader feel both righteous anger and tender concern at the same time. Content that treats empathy as a resource to be allocated, rather than a virtue to be maximised, speaks to a deep ambivalence that many educated people feel about the way their own moral intuitions are being managed. They sense, without always being able to articulate, that the version of empathy being sold to them by certain advocacy ecosystems is doing something more complicated than simply increasing net human welfare.

Another reason is generational. Research on empathic response across age cohorts — including work published in journals covering developmental and social psychology — has consistently found that self-reported empathy among younger adults has been trending in one direction for roughly two decades, while at the same time certain kinds of moral advocacy have become more vocal and more institutionalised. This is not, by itself, a contradiction. People can feel more moral emotion in the abstract while behaving in ways that look less empathic in practice — if what they mean by empathy is narrowly targeted at peers, at in-groups, at those who share their cultural reference points. The literature does not resolve whether the decline in self-reported empathy reflects genuine emotional flattening or a growing unwillingness to perform empathic response for audiences that are perceived as hostile. What it does suggest is that the relationship between feeling and acting, between moral emotion and moral behaviour, is more complicated than the folk narrative allows.

The politics of choosing not to care

What makes the toxic empathy argument politically potent is that it translates easily into a general epistemology of suspicion toward moral appeals. If empathy toward certain targets is a trap — if compassion directed at the wrong recipients is a resource extraction exercise in which the giver is the product — then the logical response is to adopt a more selective, more strategic posture toward moral obligation. The vocabulary of self-interest, long considered gauche in liberal cultural settings, regains some of its moral plausibility. It becomes not selfishness but discernment.

This is where the argument becomes contested in ways that are not easily resolved. Critics — and there are many, running from progressive commentators to clinical psychologists who study prosocial behaviour — contend that the toxic empathy frame is a sophisticatedRationalisation for doing less. The concept, they argue, gives intellectual cover to people who were already inclined to care about some groups and not others; it simply equips them with a vocabulary that makes the selectivity sound principled rather than arbitrary. The underlying behaviour — allocating moral concern based on proximity, cultural familiarity, or tribal loyalty — is what the toxic empathy framework claims to diagnose. But the framework, the critics say, does not actually change the behaviour. It changes the self-description.

Saad's defenders push back on this characterisation. The point, they argue, is not to reduce one's moral circle but to increase the quality of one's moral attention — to make the act of caring more deliberate, more consequential, and less susceptible to manipulation by actors who have learned to exploit the social prestige that attaches to visible compassion. In this reading, toxic empathy is not an argument for selfishness; it is an argument for moral seriousness. The person who chooses carefully is not the person who cares less. They are the person who cares better.

Where the debate goes from here

The viral trajectory of Saad's video reflects something genuine about the current state of moral culture in English-speaking countries. The vocabulary of empathy has become so saturated — so thoroughly colonised by advocacy, by institutional communication, by the self-presentation demands of professional public life — that the felt experience of many people is that they are being emotionally managed rather than genuinely spoken to. The toxic empathy argument offers a diagnostic label for that experience, and a posture of resistance that feels both intellectually defensible and emotionally honest.

Whether the diagnostic label corresponds to a real phenomenon, or whether it is a sophisticated folk psychology that describes something real while getting the mechanism wrong, is a question that the available research does not fully answer. What is clear is that the argument has found its moment. As institutional trust continues to erode, as social media continues to reward emotional performance over careful judgment, and as the political class continues to treat moral emotion as a resource to be deployed, the appeal of a framework that says "you do not owe your compassion to anyone who has not earned it" is unlikely to diminish. The conversation will continue. The people having it most loudly, however, are not the ones most likely to be changed by the evidence.

This publication noted the framing in Saad's video against a broader media environment that tends to treat empathy as an unambiguous good. The wire presented it as a cultural curiosity; this article treats it as a symptom.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

https://x.com/newstart_2024/status/1924356641333212160
https://en.wikipedia.org/wiki/Empathy
https://en.wikipedia.org/wiki/Reciprocal_altruism
https://pmc.ncbi.nlm.nih.gov/articles/articles/PMC5930732/