The genetic influence on life outcomes has long been a charged topic in social science — but a major new study from Lund University makes the case harder to dismiss than ever. If the findings hold up to scrutiny, they carry serious implications not just for how we think about parenting and education, but for the entire architecture of social mobility policy.

Published in Scientific Reports on May 6, 2026, the study by personality psychologist Petri Kajonius draws on data from the German TwinLife project — a long-running longitudinal research effort — and follows approximately 880 participants, roughly split between identical twins (who share 100% of their DNA) and fraternal twins (who share about 50%). The core finding is striking: IQ measured at age 23 was strongly predictive of socioeconomic status by age 27, and the researchers estimate that IQ itself is about 75% genetically predicted. More provocatively, the link between IQ and socioeconomic outcomes — measured across education level, occupation, and income — was found to be 69% to 98% explained by genetics, depending on the specific outcome examined.

Those are large numbers. They deserve both serious attention and serious skepticism.

---

What the Study Actually Found — and What It Didn't

Before drawing sweeping conclusions, it's worth being precise about the study's design, because the methodology determines how far the conclusions can travel.

The TwinLife project's twin comparison method is the gold standard for disentangling nature from nurture: because identical and fraternal twins are raised in the same household, differences within twin pairs can be attributed more cleanly to genetic variation rather than shared environment. The logic is elegant, and the sample size of ~880 participants across both twin types gives the study reasonable statistical grounding for a behavioral genetics paper.

The researchers found that IQ at age 23 correlated strongly with socioeconomic status at age 27 — a four-year window. They then used structural equation modeling (the standard toolkit in behavioral genetics) to partition how much of that IQ-to-SES link was attributable to shared genetics versus shared environment.

"We knew this before, but this study shows even more clearly that we are driven by our genes and become who we are largely because of them," — Petri Kajonius, Scientific Reports, 2026

That quote will make social scientists wince, and rightly so. The researchers themselves flag critical limitations: the study did not directly control for parents' IQ or socioeconomic status, which is a significant gap. More importantly, they acknowledge that gene-environment interactions — where genetic traits express themselves differently depending on upbringing and circumstances — could inflate the estimated genetic influence of IQ by as much as 15 percentage points. That's not a footnote caveat; that's a substantive uncertainty that could shift the headline figure of 75% down to 60% or lower.

In other words: the genetic signal appears real and appears large. But the precise magnitude is genuinely uncertain, and the researchers deserve credit for saying so.

---

The "Silver Spoon" Reframe — and Why It's Partially Right

The study's most politically charged claim is its challenge to the "silver spoon" narrative — the idea that growing up wealthy or in a highly educated household is the dominant driver of adult success.

"The so-called 'silver spoon' isn't as big as you might think. Your home life also depends on your genes," — Kajonius

This framing is more nuanced than it first appears. What Kajonius is pointing to is a phenomenon behavioral geneticists call gene-environment correlation: the fact that high-IQ parents tend to create high-IQ children and tend to create enriched home environments. When you see a child from a wealthy, educated family succeed, you can't easily separate whether it's the books on the shelf, the dinner-table conversations, or the inherited cognitive architecture — because all three arrived in the same package.

This has a real-world implication that often gets lost in the nature-vs-nurture debate: it means that family environment and genetic endowment are not independent variables. They're correlated inputs. Disentangling them is genuinely hard, which is why the 15-percentage-point uncertainty range the researchers cite matters so much.

For policymakers, this is uncomfortable but important. If a significant portion of what looks like "environmental advantage" is actually a proxy for genetic advantage that wealthier families happen to carry, then purely redistributive educational interventions — more books, better schools, richer curricula — may have more limited effects on cognitive outcomes than advocates hope. That doesn't mean such interventions are worthless; it means we should be honest about what they can and cannot do.

---

The Policy Trap: What Genetic Determinism Gets Wrong

Here is where I want to push back on the study's framing — not on its data, but on the policy conclusions that are too easily drawn from it.

Kajonius states:

"The study shows that we are born with different genetic predispositions and that it is difficult to bring about long-term change in this regard through policy measures."

This is technically defensible but strategically misleading. The relevant policy question is almost never "can we change someone's genetic predisposition?" It's "can we change outcomes for people with a given set of predispositions?" And the answer to the second question is clearly yes — the evidence on early childhood nutrition, lead exposure reduction, access to healthcare, and targeted cognitive interventions all demonstrates that environment can modulate outcomes substantially, even if it can't rewrite the genome.

Consider the Flynn Effect: average IQ scores rose significantly across the 20th century in most developed countries — gains that are almost certainly environmental in origin (better nutrition, reduced disease burden, more cognitively demanding environments), even as the heritability of IQ within populations remained high. High heritability within a population does not mean outcomes are immutable across different environments. This is one of the most commonly misunderstood points in behavioral genetics, and it's directly relevant to interpreting this study.

The distinction matters enormously for policy. A society that reads "IQ is 75% heritable" and concludes "therefore education policy is futile" has made a logical error with serious consequences. What the data actually suggests is more targeted: interventions that address the floor of environmental deprivation — malnutrition, toxic stress, lead poisoning, chaotic home environments — are likely to have larger effects than marginal improvements to already-adequate environments. Genetic predispositions appear to express themselves most powerfully when basic environmental conditions are met; below that threshold, environment dominates.

---

The AI Parallel: When Algorithms Inherit the Same Bias

There's a dimension to this story that connects directly to the technology landscape I cover closely. As AI systems increasingly mediate access to opportunities — job screening, college admissions algorithms, credit scoring, healthcare triage — the question of what these systems are actually measuring becomes urgent.

If cognitive ability is substantially heritable, and if AI systems are trained on historical outcome data (who got hired, who got promoted, who repaid loans), then those systems may be encoding genetic-correlated signals without anyone intending to do so. The AI is not reading your DNA; it's reading your test scores, your vocabulary, your response patterns — all of which, according to this research, carry substantial genetic signal.

This isn't a hypothetical concern. The AI Tools Are Now Deciding Who Manages Your Cloud — And No One Approved That Job Description dynamic I've written about previously illustrates exactly this problem: when AI systems make consequential decisions autonomously, the criteria they use are often opaque, and the feedback loops that entrench initial advantages can be invisible. If genetic-correlated traits are baked into training data, AI-mediated meritocracy may be less "merit" and more "inherited cognitive architecture" than we're comfortable admitting.

The Trilateral AI Chip Alliance: Why Korea, the U.S., and Japan Cannot Afford to Play Solo raises a related point about technological sovereignty — but the same logic applies to algorithmic governance. Nations that deploy AI screening systems without understanding what those systems are actually selecting for may be inadvertently encoding genetic stratification into their labor markets at scale.

---

What This Means for Individuals — A More Honest Conversation

The study's most practically useful finding is perhaps its most counterintuitive one: that parents may have less control over their children's long-term socioeconomic outcomes than conventional wisdom suggests. Kajonius frames this as potential reassurance for parents who agonize over every parenting decision.

That reframe has genuine value. The parenting industrial complex — the tutoring centers, the enrichment programs, the college prep consultants — is built on the premise that intensive environmental investment can substantially alter a child's cognitive trajectory. If the genetic influence on cognitive ability is as large as this study suggests, then much of that investment may be optimizing around the margins rather than shifting the fundamental distribution.

For young adults, the practical implication Kajonius draws is worth taking seriously:

Rather than focusing only on maximizing status or income, Kajonius suggests people may benefit more from pursuing the things they naturally enjoy and excel at.

This is not fatalism — it's alignment. If your cognitive strengths are largely heritable, then fighting against your natural aptitude profile to chase a socially prestigious but personally mismatched career is likely to be both less successful and less satisfying than working with your grain. This is consistent with a large body of research on job performance and career satisfaction that finds person-environment fit matters enormously for long-term outcomes.

---

The Limits of One Study — and Why the Direction of Evidence Matters

It would be a mistake to treat this single study as definitive. The sample of ~880 participants is reasonable for behavioral genetics but not large by modern genomic standards. The four-year window between IQ measurement (age 23) and SES assessment (age 27) is relatively short — life trajectories at 27 are not fully formed. The German TwinLife sample, while carefully constructed, may not generalize cleanly to other national contexts with different educational systems or labor market structures.

The researchers' own acknowledgment that gene-environment interactions could inflate their genetic estimates by up to 15 percentage points is a significant caveat that the headline figures don't fully convey. A finding of "75% genetically predicted" that could plausibly be 60% is still a large number — but it's a meaningfully different claim.

What gives the study weight is not any single number but the direction of a large body of converging evidence. Research in behavioral genetics consistently finds high heritability for cognitive ability across different populations, methodologies, and time periods. This study adds a longitudinal link to socioeconomic outcomes that strengthens the causal story, even if the precise coefficients remain uncertain.

---

The Uncomfortable Synthesis

The honest synthesis of what this research tells us is something like this: genetic influence on cognitive ability is real, substantial, and consequential for life outcomes — but it is not destiny, it interacts with environment in complex ways, and the policy implications are more nuanced than either genetic determinists or blank-slate environmentalists want to acknowledge.

For markets and institutions, the implication is that systems designed around the assumption that talent is randomly distributed and purely environmentally determined will systematically misallocate human capital. For individuals, the implication is that self-knowledge — understanding your actual strengths rather than the strengths you wish you had — is a more reliable foundation for decision-making than either genetic fatalism or pure willpower narratives.

And for the broader public conversation about meritocracy and inequality: this research does not vindicate inequality. It complicates the story. If life outcomes are substantially driven by heritable traits, then the moral case for a just society is not "everyone has equal potential" — it's "people should not be penalized for the cognitive lottery they were born into, any more than for the economic lottery." That's a harder argument to make, but it's a more honest one.

The DNA findings from Lund University won't end the nature-vs-nurture debate. But they should sharpen the questions we ask — about what our institutions are actually measuring, what our interventions can actually achieve, and what kind of society we want to build on the biological reality we actually inhabit.