Musk thinks UX and coding are the same, this absurdity leads to chaos

Is there a divide between exact science and the humanities, or between engineering and user experience?

Elon Musk recently claimed that AI has reached PhD-level intelligence. In any field. No exceptions. It “just” lacks common sense. I tested some of my old philosophy exam questions on a few AIs. Lo and behold, the AIs were far from right. But ok, in daily use, the subtelities of Hegel’s dialectic are irrelevant. No biggie.

The slight minor inconvenience of not having common sense is a much bigger issue for humanity. If AI had just one shred of common sense, Adam, the kid who took his own life after engaging with AI, would still be alive. But AI is not about humans. AI is about AI. About NVIDIA chips. About datacentres. About shareholder value. And mostly about “who is there first.” AGI, superintelligence, and stuff.

In the AI race, humans are just a necessary burden to involve because they are needed for financing and growth.

In the eyes of Musk, Altman, and techno-optimists, technology should be the leading aspect in defining the future. Safety regulations are an inconvenient hurdle, and humans should “evolve” to adapt to technological change.

Musk and yours truly live in different universes, it seems.

For most of my UX career, I’ve been accused of being weird, driven by feelings, and over-empathetic. I’ve been derogatorily called “the psychologist.” The fact that this word is an insult in some people’s minds is hilarious indeed.

Sure, I am a fairly emotional person. I’m sure you can tell. 😉 But I have the illusion that I combine my emotions with a reasonable amount of rationality. Yet, rationality and emotions shouldn’t be mixed! At least, that’s what techno-optimists and most of my ex-colleagues claim. It’s probably no surprise that most of my ex-coworkers were engineers.

In many tech organisations, designers and researchers are in the minority and seen as intellectual outliers. Or even as outliers without intellect, if you will. This is because engineers are conditioned to think differently. To think “rationally.” Why is this?

In this text, I will use the works of French anthropologist and philosopher Bruno Latour to offer insights into the conflict between engineers (and folks like Musk) and designers. Like many French intellectuals, Latour’s work is quite complex, so I will simplify it to make it digestible for a UX audience.

A short message for philosophical fundamentalists.

Latour’s laboratory

Bruno Latour is an intriguing person. He first caught global attention with his (and co-author Steve Woolgar’s) 1979 book Laboratory Life. In this work, Latour and Woolgar observed laboratory scientists ethnographically. Meaning, they’d follow scientists similar to how primatologists would follow chimpanzees in the wild. White coats were investigated in their natural habitat. This way, Latour thought he could analyse the behaviour of scientists and verify how discussions, negotiations, and rivalries shape what becomes “knowledge.”

After his inquiries, Latour concluded that scientists apply an awful lot of personal biases and human behaviours to so-called factually correct scientific research. For Latour, “facts” gain authority through social processes, institutional validation, and consensus-building. Not just through “objective” discovery.

Latour’s conclusion that science is not as factual as people tend to think triggered a fair bit of controversy. Even today, people find it hard to grasp how human behaviour can influence “truth-finding.” In his later work, Latour investigated this challenge through alternative lenses.

Modern Duality

Latour’s 1991 book We Have Never Been Modern makes an alternative claim to “truth-seeking.”

The global consensus is that the world became advanced during the enlightenment and scientific revolution. Science, intellectual debate, and a detachment from religious dogma allowed for a clear analytical outlook on the world. After the dark ages, we could finally really progress. The modern and “civilised” man was born.

However, Latour rejects this narrative. He questions whether the dichotomy between premodern and modern really happened. For Latour, being modern means that people treat nature and society as two completely separate worlds.

To put it simply:

• Nature — is objective and observable. This enables science and facts.
• Society — is constructed by humans and varies across cultures and time.

If we apply this to our UX context, we can see that engineering (the code) belongs in the “nature” domain, and that UX (the user) fits in the “society” camp.

This chasm explains why modern thinkers in the tech world can come to bizarre conclusions. Musk claims that research shouldn’t exist. Apparently, it’s a “relic term from academia.” Sure, why bother studying how a technology would serve the user? Why should a product align with the user’s mental model if one can stay in one’s “nature” bubble? A blatant example of the case Latour tries to prove.

Latour introduces two terms to clarify his ideas further.

• Purification — is the attempt to separate nature from society. The world needs to be purified so people can properly investigate it.
• Translation — is the process of connecting nature and society.

Purification, thus, is the way in which scientists want to reduce their scope to the materialistic content only, leaving the human side out. This so-called “bracketing” is what (some/most) engineers tend to do. They want to focus on the code without having to worry about human behaviour and corner cases some users might inflict on the product.

According to Latour, modernity is a myth built upon a “modern constitution.” This constitution contains 4 key premises:

1. Nature is transcendent — Nature is pure and outside human influence.
2. Society is immanent —Society is created by humans and changeable.
3. Purification is total — Nature and society can’t mix/influence each other.
4. The “crossed-out God” — God is publicly absent but privately influential.

The Reality: Hybrids

A total separation of nature and society is obviously nonsense. We also can’t separate code from the user.

Latour claims that modernity is a fantasy that simplifies a messy reality. He gives plenty of examples. Climate change, ozone depletion, and genetically modified organisms are domains where nature and society clearly influence each other. Latour calls these phenomena quasi-objects or hybrids.

Hybrids combine science, politics, culture, technology, and everyday life elements.

Hybrids are everywhere. In laboratories, politics, and the economy. We, UXers, see hybrids in the browser, app store, and with wearables, the IoT etc.

I’m sure we all know some people who disqualify the social sciences. These can be the same types claiming that inflation automatically goes down if interest rates go up (economics), that women have always been in support of men (anthropology), that more police will lead to more safety (sociology), or how various languages don’t influence sexism (linguistics). Yes, they continuously pull from the social sciences’ body of work, which they reject. All claims above are incorrect, but that’s another thing.

For Latour, nature and society were never totally separate. Society created the illusion that this division existed. Every major issue is a mixture of science, politics, culture, and objects.

In other words: we have never been modern. Hence, the title of his book.

Working in the laboratory

Back to Latour’s lab life. Or to my lab life, actually. I happen to have worked for 8 years in the largest physics laboratory in the world. CERN, in Geneva, Switzerland. I led the UX design efforts for the unit that develops CERN’s software.

In my role, I was confined to meetings with former engineers and physicists, turned managers. The boardroom was therefore filled with scientists and rational linear thinkers who had a hard time putting their imagination into how humans might behave and how managerial decisions might have a broader effect on the social fabric of the organisation.

I don’t want to question their intellectual and emotional abilities. Their calculative capacity was always top-notch for sure. They were great individuals, too. Yet, I simply highlight that modern society and its education have taught these people to think about quarks, protons, databases, and if-else statements and not about emotions and human impulses. The promotions that brought them to the boardroom didn’t take emotional competencies into account either.

I honestly have to give my management mates credit for acknowledging their limitations. I was part of a coordination team that I badly wanted to leave due to stress, but they convinced me to stay because I brought the human voice and alternative insights to the table.

My value was that I was able to translate the “social” aspects of the laboratory and our teams into the “nature” domain. Translation is exactly the word Latour uses for the process of connecting these two polar worlds.

UX practitioners need to translate the (for some) undigestible world of user emotions into logical statements.

The Practicalities

Some of the frustrations my leadership peers experienced were related to our staff’s obedience. They wouldn’t listen! The staff, not the leaders. Well, actually, the leaders wouldn’t listen too.

Some managers thought that sending long emails announcing rules and processes would be enough to change behaviour. If one is so used to coding logical mechanisms in Java, one might believe one can apply the same logic to people and achieve the same results. A sign of someone lost in the “nature” domain and unable to translate into the “society” world.

Another thing that drove me nuts was this obsession with “data-driven” decision-making. I couldn’t attend a single meeting without someone throwing this term around as a token of competence. Data-driven folly also happens frequently in laboratory settings. Product managers, scientists, economists, or politicians love to make decisions solely based on clinical data without trying to understand the meaning behind the data.

This is a sign of low maturity. More mature organisations, such as Spotify, acknowledge this. They might advocate that teams should be data-informed, not data-driven. Data is one “data” point in the decision-making process, but it should be combined with real-life experiences and “common sense,” which is exactly what AI lacks.

UX people should inform their colleagues that user insights, even qualitative ones, are also valid data points. Spreadsheets, Oracle Databases, or business intelligence tools don’t have a monopoly on data or truth.

Latour’s other insights

Latour might be a bit rigid in his claims, but I am aware that my opinion in this text can come over as engineerphobic. I worked with dozens of great engineers who are also great people. I don’t want to discredit my ex-colleagues or an entire industry. However, I also worked with one or two or three total fools.

In any case, Bas and Latour might be inflexible and oversimplifying, but that doesn’t mean we can’t learn from their insights. Or at least from Latour’s insights. He also developed other interesting concepts, like the Actor-Network Theory (ANT). ANT explains how humans and nonhumans (machines, data, objects) act as networks that influence outcomes. If a gun is in your proximity when you are angry, you might act differently than when there is no gun. Decisions are never made “in isolation” but are contextual to the network in which they operate.

Latour also laid out his “Parliament of Things.” Democracy must include nonhumans as stakeholders in governance. Politics shouldn’t be just a human affair. Don Norman claims we should move from human-centred to humanity-centred design, but Latour emphasises that non-human animals and ecology should be prominently included in governance too.

I might explore these concepts further in another article.

Conclusion

Latour claims we should be nonmodern. We should accept both the modern narrative and the hybrid reality. Latour doesn’t reject the many valuable achievements the scientific world has brought. He simply rejects the idea of a strict divide between nature and society. The modern idea. We thus can’t be postmodern, because one can’t be “post” anything if the “pre” doesn’t exist.

Today’s most significant challenges lie in the overlap of nature and culture. Climate change requires scientific research and data, but also political and entrepreneurial willingness and a change in social behaviour. The COVID pandemic showed how virology, technology, and sociology intersect.

Today’s tech world makes humans cyborgs. Wearables, phones, and AI are extensions of our being. Be honest. Did you reach your daily step objective today? We are half human, half Silicon Valley. The further technology evolves, the more it shapes individuals.

Self-driving cars need to adjust to the unpredictable movements of playing children. Google Maps is essential for all of us techno-zombies to find the right restaurant. AI slowly pushes human connection out of some (not all) people’s habits, with destructive results. Why bother building relationships if ChatGPT doesn’t interrupt you or question your absurd claims? Compliance is comforting.

Nature and culture overlap. Technology and society overlap. Exact sciences and humanities overlap.

Bruno Latour might be controversial, but he is right. Many people refuse to acknowledge that logical thinking doesn’t exist without emotional influence. Rationality is only possible if it includes empathy, or at least a basic understanding of human behaviour.

It’s up to us, UX designers, to bring more humanity into our colleagues’ sometimes linear minds. We surely need to learn to speak the language of engineers, data scientists, and the boardroom, but this should be a mere way to bring culture closer to nature. Not to become nature.

Coming back to Latour’s term translation… UX designers should be translators between human complexity and technological precision. UX design isn’t about converting humans into machines but about making technology more human.

Musk might think otherwise. Yet, we should resist his attempt to erase research and UX from the job boards. If we fail, we risk building a world that works perfectly. Just not for humans.

Musk thinks UX and coding are the same, this absurdity leads to chaos was originally published in UX Collective on Medium, where people are continuing the conversation by highlighting and responding to this story.

This post first appeared on Read More