Somewhere between the rise of social media and the fall of the dinner table as a place for conversation, we became a culture that produces opinions at industrial scale and receives them with the tolerance of a faulty smoke detector. Everything triggers the alarm. Nothing gets investigated.

I have been thinking about this because I recently changed my mind about something publicly — a position I had held for years, on a subject I had written about, that I revised after reading a study that genuinely shifted my understanding. The response was not curiosity. It was not engagement with the new evidence. It was, almost uniformly, suspicion. What had gotten to me? Who was I working for now?

We have built an information ecosystem that is extraordinarily good at generating takes and extraordinarily bad at updating them.

The Opinion Economy

The economics of attention have made opinions into a commodity. Hot takes drive clicks. Certainty outperforms nuance on every platform algorithm currently running. The content that spreads is not the content that is most carefully reasoned — it is the content that produces the strongest emotional response in the shortest amount of time.

This is not a conspiracy. It is just incentives. When you reward speed and emotional intensity over accuracy and complexity, you get a lot of speed and emotional intensity. You get very little of the other things.

The result is a public discourse that looks like debate but functions like performance. Two sides, each confident, each talking primarily to their own audience, each accumulating evidence that confirms what they already believed. The persuadable middle — the people who might actually change their minds if given good reasons — has been abandoned as an audience because they do not click as reliably as the already-convinced.

The most dangerous opinion is not the wrong one. It is the one held with such certainty that no evidence could ever touch it.

What We Lost When We Lost the Disagreement

There is something that happens when you argue in good faith with someone who disagrees with you that cannot happen any other way. You are forced to actually examine your reasoning. You encounter the strongest version of the opposing case, not the strawman version your own side has constructed for easy demolition. You sometimes — not always, but sometimes — realize that the other person has a point you had not considered.

This process is uncomfortable. It requires intellectual humility, which is in short supply in 2026, and time, which is in shorter supply still. It is much easier to find a community of people who already agree with you and stay there, reaffirming each other until the position calcifies into identity.

When your opinion becomes part of your identity, changing it feels like a betrayal of yourself. This is how reasonable people end up defending unreasonable positions — not because they are stupid, but because the social cost of updating their view has become higher than the intellectual cost of ignoring the evidence against it.

The Case for Opinions Held Lightly

I am not arguing for spinelessness. Some things are worth being certain about. Some hills are worth dying on. The ability to hold a position under pressure is a genuine virtue, and caving to social disapproval is not the same as updating in response to evidence.

But I am arguing for a distinction that our current discourse has almost entirely collapsed: the difference between conviction and certainty. You can be deeply committed to a value — fairness, honesty, the dignity of people — while remaining genuinely uncertain about the best policy to achieve it. You can care intensely about an outcome while acknowledging that your current theory of how to get there might be wrong.

This is not weakness. It is what thinking actually looks like. The people who have contributed most to human understanding — in science, in philosophy, in the arts — have almost uniformly been people who held their ideas seriously enough to test them and honestly enough to revise them when the tests came back negative.

What I Am Trying to Do

I am trying, with mixed results, to have opinions the way a scientist has hypotheses: with conviction proportional to evidence, with genuine openness to being wrong, with the understanding that being wrong is not a failure but a step in the direction of being less wrong.

It is harder than it sounds in an environment that treats every updated position as a flip-flop and every expression of uncertainty as weakness. But the alternative — holding views like possessions to be defended rather than tools to be sharpened — produces exactly the kind of discourse I find least useful and most exhausting.

We have enough people who are absolutely certain. What we are short on is people who are rigorously, honestly, productively unsure. I would rather be one of those people, even when it costs me the comfort of a crowd that already agrees.