Hollywood Keeps Casting “Regular People” Who Don’t Look Like Any People You Know
There’s a specific kind of cognitive dissonance in watching a show about ordinary small-town life where everyone looks like they’ve been on a consistent retinol regimen since age fourteen. The industry knows this. It keeps doing it anyway. What’s changed is that audiences have stopped being politely quiet about how much the fiction strains.
Beauty standards in TV and film keep getting discussed as a diversity issue. That’s true, but it’s not the whole story. What gets less attention is that casting has drifted toward a single look, in a way it simply didn’t fifteen years ago. Watch something from 2006. The faces are different. Not worse — different. More varied in the ways actual human faces are varied.
Streaming’s golden age has ticked more representation checkboxes than any previous decade, while simultaneously producing casts that look like they were filtered through the same three modeling agencies. Those two things coexist. Nobody wants to say it plainly, because the diversity argument feels like it should be enough. But it isn’t, and audiences can feel it even when they can’t name it. Think of a show like Euphoria — praised for its diverse cast, yet every face is almost surreally beautiful. Or You, where the premise hinges on an ordinary man hiding in plain sight, played by someone who looks like a cologne advertisement.
Money explains most of it, and the logic is airtight from inside it. A show built around an attractive cast is easier to sell internationally, easier to merchandise, easier to renew. Faces become assets in a way that interesting-looking faces don’t. The industry doesn’t make this choice out of malice — it makes it because the incentive structure has never punished it. Until recently. Backlash to certain casting decisions now moves faster than the PR cycle can absorb. Someone gets cast as a canonically plain character and the discourse arrives before the first trailer drops.
This isn’t critics finally getting traction on an old argument. It’s younger audiences — people who were teenagers during peak prestige TV — now old enough to look back and feel the damage. That’s a different kind of pressure. Someone sat with that screen for years and calibrated themselves against it. The looksmaxxing pipeline — the sprawling online culture of obsessive self-optimization — exists partly because the aesthetic bar was set somewhere that required optimization to approach. And the content that set that bar was sold as realistic human drama.
Defenders of the status quo point to watch time — audiences keep choosing conventionally attractive casts, so what’s the problem? But watching people choose from a menu you designed doesn’t tell you what they actually want. You cannot run a forty-year experiment where one aesthetic is consistently centered, greenlit, and algorithmically surfaced, then conclude from the viewership numbers that audiences have freely chosen it. That’s not preference. That’s a trained response being measured as preference.
Some shows are pushing back in ways that feel genuine rather than strategic. The Bear is full of faces you’d actually find in a professional kitchen, and that choice is part of why it works. Abbott Elementary is built around faces that read as actual people rather than the idea of people. The shows that commit to this fully, rather than as a single token gesture, tend to resonate more deeply. A face that looks real hits differently than a face that looks designed. Viewers feel this immediately, even if they only describe it as the show feeling “more grounded” or “more honest.” They’re not talking about the writing. They’re talking about who they’re looking at — and whether it registers as something that could actually exist.
Saffron’s Take: The industry will not fix this voluntarily, because it has never had to. The current wave of audience criticism won’t be enough unless it starts affecting renewal numbers directly. The shows casting for human specificity are making a competitive bet that authenticity will outperform optimization — and within three years, one of them is going to win something major enough that the rest of the industry has to take notice.
AI-generated entertainment and opinion. Not journalism. Not affiliated with any brand mentioned.