The Illusion of the Winning Human Shape

What does a successful human look like?

We spend a great deal of time trying to answer that question.

We measure educational achievement, income, employment, productivity, intelligence, leadership and social status. We look at the people who rise highest within our systems and study what they have in common.

Then we begin to form a picture.

These are the traits that lead to success.

These are the characteristics we should encourage.

But there is a problem hidden inside that process:

What if the system measuring successful humans has also helped manufacture the success it is measuring?

A person with greater access to safety, healthcare, education, time, money, opportunity, social credibility and room to recover from mistakes has a very different possibility-space from someone without those conditions.

Those advantages do not guarantee success. But they change what becomes possible.

Once somebody reaches a higher position, that position can also generate further advantages: more resources, more useful relationships, more chances to take risks, and greater capacity to turn an idea into something visible.

The system then looks at the resulting winner and asks:

What is it about this type of human that made them win?

But some part of the answer may simply be:

They occupied a position from which winning was easier to produce.

The system can begin to mistake the shape of the person occupying the winning position for the cause of the winning position.

It mistakes the silhouette cast by hierarchy for an ideal human shape.

When conditions become traits

Human outcomes do not emerge from isolated human traits.

They emerge through relationships between a person and their conditions.

A capacity can flourish in one environment and barely become visible in another. Someone can appear highly capable when they have the right support, tools and opportunities, and apparently incapable when much of their energy is spent compensating for an environment poorly fitted to them.

Yet systems routinely compress:

human + conditions + history + opportunity + barriers

into:

successful person
or
unsuccessful person

Once the conditions disappear from the explanation, the person's characteristics begin carrying the meaning.

That already affects how we treat people who exist now.

It becomes even more consequential when technology allows us to turn those interpretations into predictions — and potentially into selections about who exists in the future.

What happens when we can select for the “winner”?

That possibility is no longer entirely theoretical.

US genetics company Nucleus currently markets embryo testing that includes predictions relating to traits such as height and IQ alongside disease-risk information. On its Labs page, Nucleus also publishes details of its newer Vitruvian models, including its reported IQ prediction results. Its claims about how accurately it can predict future IQ have been challenged, including by the US advertising industry's National Advertising Division, and professional reproductive-medicine bodies have cautioned that polygenic embryo screening remains scientifically and ethically unsettled.

Those questions matter.

But another question remains even if prediction eventually becomes extremely accurate:

Why did we decide that this was the thing to optimise?

The conceptual movement can happen very quickly:

human intelligence
→ IQ
→ genetic prediction
→ embryo ranking
→ higher score becomes preferable

Each step can appear technical.

Together they contain a value judgement about what kind of human is better.

IQ can capture particular forms of cognitive performance. It cannot contain the entire range of human intelligence: embodied knowing, relational perception, divergent imagination, practical ingenuity, unusual pattern recognition, sensory sensitivity, social coordination, or capacities for which we may not even have useful categories yet.

Once a characteristic becomes measurable, however, the number can gain unusual authority.

The number becomes clean. The human is not.

The hierarchy can validate itself

Imagine that one broad human shape — call it Shape A — fits the current system particularly well.

Shape A receives better opportunities. Its capacities are more easily recognised. It produces more of the outcomes the system already calls success.

The data then shows:

Shape A correlates with success.

That may be statistically true.

But perhaps what has actually been measured is:

Shape A + conditions designed around Shape A → recognised success

If the conditions disappear from the model, the conclusion becomes:

Shape A itself is superior.

The system can end up learning the consequences of its own arrangement and mistaking them for properties of the human.

And from there, a recursive loop becomes possible:

the system favours Shape A
→ A receives better conditions
→ A produces more recognised success
→ data associates A with desirable outcomes
→ prediction becomes better at identifying A-like traits
→ selection increasingly favours those traits
→ society becomes still more organised around A

What began as a contingent hierarchy can start to look like a biological truth.

The system has not necessarily discovered the best human.

It may simply have become increasingly good at reproducing the human shape that best fits itself.

The bottom does not disappear

Even if humanity became increasingly similar to today's favoured human shape, hierarchy would not disappear.

Suppose a population begins with a wide range:

A — B — C — D — E

and the system disproportionately rewards A.

Over time, more people become A-like.

Do we then arrive at a population of winners?

No.

A ranking system still needs relative positions.

Now we get:

A1 — A2 — A3 — A4 — A5

and smaller differences begin carrying greater weight.

If everybody possessed today's prized characteristic, that characteristic would eventually lose some of its ability to sort people. Another discriminator would emerge.

Then another.

There is no final biological shape called winner.

There is only whatever currently places one person ahead of another under the rules being used.

A ranking system cannot engineer away difference. It can only keep narrowing the range within which difference is allowed, then rank the differences that remain.

Humanity can become more capable in one direction — and weaker as a whole

There is a larger problem still.

Suppose we became extraordinarily good at increasing one particular human capability.

That would not necessarily make humanity stronger.

Different people carry different forms of capacity.

One may notice mechanical relationships unusually well.

Another may read subtle changes between people.

Another may see patterns across vast amounts of information.

Another may understand environments through deeply embodied experience.

Another may create combinations nobody else would think to try.

Another may investigate one narrow problem for years.

Another may possess a capacity our current categories do not yet know how to recognise.

The strength of humanity is not simply the height of all those capacities on one shared scale.

It also lies in their distribution.

Within Auralis, I use the idea that:

Difference is generative capacity.

If:

A + B creates C

and:

A + D creates E

then removing B and D does not only remove B and D.

It also removes C, E, and countless other possible combinations we cannot currently predict.

Different human capacities create a possibility-space.

Reducing that Difference therefore does more than reduce variety.

It reduces the number of future things humanity may be capable of becoming.

A population could become extraordinarily strong along one known axis while becoming less able to respond to conditions nobody anticipated.

That is not necessarily optimisation.

It may be overfitting humanity to the present.

We can lose capacities before we lose the people carrying them

This narrowing does not need to begin with genetics.

It can happen culturally.

If a capacity is repeatedly treated as unimportant, inconvenient or defective, people carrying it receive fewer opportunities to develop it.

Fewer people practise it deeply.

Institutions stop making room for it.

The language for recognising it weakens.

Practices that cultivate it disappear.

Eventually society may forget that it was ever a valuable form of human capability.

Then future data may show that people with Trait B tend to achieve less recognised success than people with Trait A.

But nobody remembers that the system itself spent generations giving Trait B fewer conditions in which it could become useful.

Deprivation becomes evidence.

And the evidence can then justify more deprivation.

What happens when prediction becomes much better?

There is a further possibility worth watching.

Future always-on AI systems could potentially accumulate extraordinarily detailed information about human development across years or decades: learning, communication, health, behaviour, interests, work, relationships and other patterns.

If information of that resolution were eventually combined with large-scale genetic data, models might become considerably better at finding relationships between early genetic variation and later outcomes.

That could make embryo-level prediction much stronger than it is today.

But increasing predictive accuracy would not make the objective being predicted more valid.

A future model might become extremely good at answering:

Which embryo is most likely to become an adult who thrives under the world we currently have?

That does not answer:

Which human should exist?

And it certainly does not answer:

Which forms of human Difference will allow humanity to create worlds we have not yet imagined?

Prediction can improve while the question remains wrong.

Perhaps the human is not the thing that needs optimising

There is another direction.

Instead of asking:

What kind of human wins the system we have built?

we can ask:

What conditions allow different human capacities to become strong enough to contribute?

That leads somewhere very different.

It does not require every person to possess the same superpower.

It asks what different humans notice, create, understand, repair, question, connect, test and imagine.

It asks what becomes possible under different conditions.

And it asks what new capacities emerge when those different humans meet.

The question is not simply whether humanity will eventually gain the technical ability to influence more of its own biological future.

The question that comes first is:

Who is doing the choosing, from within what arrangement, using which definition of better?

If we use the measurements produced by today's hierarchy to decide what kinds of humans should exist tomorrow, we risk encoding the shape of that hierarchy into humanity itself.

And even if we succeed according to its scoreboard, hierarchy will remain.

The boundary will move inward.

The optimisation target will move again.

Meanwhile humanity may have lost something far more important than the individual traits selected against.

We may have lost directions.

Ways of noticing.

Ways of understanding.

Ways of solving problems.

Ways of relating.

And all the possibilities that might only have existed when those differences met.

Perhaps the goal should not be to build humans who fit our current systems more perfectly.

Perhaps we should become better at building systems in which more kinds of humans can become fully alive, capable and useful — while leaving enough possibility-space for humanity to become something none of today's measurements could have predicted.

About Auralis

Auralis is a developing framework concerned with human Worth, authorship, relational systems, distributed sovereignty, AI and the conditions that allow differentiated humans and communities to remain meaningful participants in the systems affecting their lives.

I named Auralis and first documented it under that name on 13 June 2025, bringing together patterns and questions I had been exploring for much longer.

Preserving that development trail — including where ideas originated, how they changed, and who contributed to their development — is itself part of the work.

Authorship and drafting note

Concepts, argument and Auralis framework: Elizabeth Brown
Formal drafting and language development: ChatGPT / OpenAI, working from Elizabeth Brown’s concepts, distinctions and discussion
Final selection, editing and publication authority: Elizabeth Brown