Stanford research psychologist Jamil Zaki has suggested that AI may be deepening what he calls a crisis of connection, according to McKinsey Insights. His point, as McKinsey Insights frames it, is about leadership and empathy inside organizations managing change, not about marketing at all. But it raises a question that sits close to the center of branding right now: if AI is making people feel more distant from each other, what happens when brands lean on AI to speak to those same people?
The temptation is obvious. AI tools can draft copy, generate images, and personalize messages at a scale no team of writers could match. For a business trying to stay visible across several channels at once, that's a real advantage. Nobody is arguing against efficiency, but usefulness alone may not be enough to hold an audience's attention, let alone its trust.
But efficiency and connection are not the same thing, and the gap between them is where a lot of brand communication quietly fails. A perfectly polished caption, written in seconds and indistinguishable from a hundred others in the same feed, might tick the box of consistency. It rarely makes anyone feel understood. Consistency tends to keep a brand visible, but it doesn't on its own make anyone care.
That's the part of Zaki's argument, as McKinsey Insights describes it, that translates well beyond the workplace. Empathy, in his framing, is what allows trust to build. Brands that want to be trusted face a version of the same test. Audiences aren't evaluating a brand's output for grammar or polish. They're reading it for signs that someone actually considered them before pressing publish.
This is where the case for human-sounding brands gets interesting, and where it also gets oversold. Imperfection on its own isn't a strategy. A typo doesn't make a brand more relatable, and an overly casual tone can feel just as hollow as a corporate one if it's not backed by anything real underneath. The brands that come across as human tend to be the ones willing to say something specific, admit a limitation, or respond to a customer in a way that wasn't scripted for every customer. That takes more effort than automation, not less.
There's also a question of scale that deserves honesty. A small business can personalize every reply because it has a handful of conversations to manage. A larger brand, running campaigns across markets and languages, cannot write everything by hand, and pretending otherwise isn't realistic advice for most organizations. The practical answer probably sits somewhere in the middle: let AI handle the volume, and keep human judgment where the stakes are highest, in the tone of an apology, the wording of a sensitive announcement, the moments where a customer is frustrated rather than merely curious. Where a brand draws that line says more about its priorities than any tagline could.
I'd argue the mistake isn't using AI in brand communication. It's assuming that output volume and audience connection move together. They don't, and treating them as the same thing is how a brand ends up technically present everywhere and meaningfully felt nowhere.
This distinction carries its own weight for any brand writing across more than one language. A message that's grammatically correct but emotionally flat will read as flat no matter where it lands. Getting the words right is necessary, but getting the intention behind them right is what audiences actually notice, and that's harder to check with a spellcheck or a style guide.
None of this means brands should treat AI with suspicion or step back from using it well. The tools are only going to get better at producing convincing, fast, well-formatted content. That's precisely why the brands that stand out may be the ones willing to slow down at the exact points where speed stops helping, and ask whether what they're about to publish sounds like it came from someone who was paying attention.
Sources
- What AI can’t replace—and why it matters for value — McKinsey Insights, 27 Oct 2026
Useful to someone on your team?