The Unoptimised Human — What the Algorithm Can't Reach

July was the month I kept writing about the same thing from different angles without fully realising it. Every piece circled the same tension: AI is optimising everything it can touch, and the things it can’t touch are the ones that actually matter.

The algorithm can process data at a speed that makes human analysis look quaint. It can segment audiences, generate copy, predict behaviour, and automate decisions that used to require a team. What it cannot do is feel. It cannot belong. It cannot understand why a person chooses a Korean luxury car they’ve never seen over a Japanese brand their family has trusted for thirty years. It cannot grasp why a bank employee doesn’t just “update their CV and move on” after being told a model is replacing them.

This month’s pieces all orbit that gap, the space between what AI optimises and what humans actually need. If you read only one, read the “Competent Beige” piece. But read them together and a pattern emerges that I think matters for anyone working in marketing, leadership, or the increasingly blurry space between humans and machines.

The Pieces

“Competent Beige”: When AI Optimises the Soul Out of Your Brand 10 July 2026

This one hit a nerve. The argument is simple: when you let AI optimise your brand messaging, you get something technically correct and emotionally dead. I called it “Competent Beige” because that’s exactly what it looks like. Polished, professional, and completely forgettable. The piece explores why algorithmic optimisation gravitates towards the mean, why that mean is the enemy of distinctiveness, and what happens to brands that confuse fluency with personality. If your brand sounds like every other brand using the same tools, you haven’t been optimised. You’ve been averaged.

The Third Need: Why Lawrence of Arabia Could Win a War But Not His Own Mind 6 July 2026

T. E. Lawrence had autonomy. He had competence. What he never found was belonging. This piece uses his life to explore the most neglected of the three psychological needs from Self-Determination Theory: relatedness. Lawrence could unify warring tribes and dismantle Ottoman railways, but he couldn’t find a place where he simply belonged. The connection to the AI conversation is this: organisations are optimising away relatedness and calling it efficiency. Every redundancy announcement framed as “investing in AI” is, at its core, a relatedness destruction event. The people who stay don’t feel relieved. They feel conditional.

When Your Bank Replaces You With an Algorithm: The Starling Problem 1 July 2026

Starling Bank announced 130 job cuts explicitly to redirect savings into AI investment. The framing was business rationale. The psychological reality was something else entirely. This piece examines what happens when the psychological contract between employer and employee gets not just renegotiated but deleted. Self-Determination Theory predicts the outcome with uncomfortable accuracy: when autonomy, competence, and relatedness are simultaneously threatened, people don’t quietly adapt. They disengage. The piece also explores why Dual Process Theory means most people never make it past the initial threat response to strategic thinking, and why leaders who don’t understand the Neuroticism distribution in their workforce will misread the transition entirely.

Your Customers Don’t Have Segments — They Have Attention Signatures 28 July 2026

The closing piece of the month reframes how we think about consumer understanding. Traditional segmentation puts people in boxes. Attention signatures recognise that the same person behaves differently depending on context, cognitive load, and emotional state. The argument is that AI-driven segmentation is getting more precise about the wrong thing. It’s optimising for patterns in behaviour while missing the psychology underneath. If you want to reach people, stop asking “which segment are they in?” and start asking “what’s capturing their attention right now, and why?”

What I’m Watching

The EU AI Act deadline has arrived. August 2nd was the compliance deadline for prohibited AI practices under the EU AI Act. Social scoring systems, emotion recognition in workplaces and schools, and untargeted facial recognition scraping are now formally banned in the EU. The enforcement mechanisms are real: fines up to €35 million or 7% of global turnover.

What interests me isn’t the regulation itself. It’s the psychological signal it sends. The EU is essentially saying: there are things AI should not optimise. Emotions. Identity. Social worth. The algorithm stops where the soul begins. Whether organisations actually internalise that boundary, or just build more sophisticated workarounds, is the question that will define the next twelve months.

The Question

Here’s what I’ve been sitting with all month. Every piece I wrote was, in some form, asking the same question:

What happens to the things that make us human when we optimise everything around them?

Brand distinctiveness gets averaged into competent beige. Employee belonging gets cut as a line item. Consumer identity gets reduced to a segment. Relatedness gets treated as a soft skill rather than a structural need.

The algorithm is extraordinary at what it does. But what it does is find the mean, reduce the variance, and optimise for the measurable. The things that actually drive human behaviour, belonging, identity, emotion, the feeling of being seen, these are not measurable. They are not optimisable. They are, by definition, unoptimised.

So here’s the provocation: Is your organisation optimising for the things that matter, or optimising away the things that matter because they can’t be measured?

I’d genuinely like to know. Because I think the answer determines whether AI makes us better or just makes us blander.

David Chadderton spent his twenties and thirties teaching people how to make life-or-death decisions at forty thousand feet. He now applies the same principles to consumer psychology, which, depending on the brief, can feel equally high-stakes. He’s the creator of the STAR Framework and the author of The STAR Framework: Rewriting the Rules of Consumer Engagement (NYC Big Book Award 2025), The STAR Operating System: Decode Mindset, Understand Motivation, Transform Human Behaviour, and Dear Algorithm, It’s Not Me, It’s You. By day, a Chief Marketing Officer. By night, a behavioural science obsessive who writes The Unoptimised Human because he can’t stop thinking about why people do what they do.