Every few months, there is a new story about social media algorithms causing harm. Algorithms that radicalise users. Algorithms that spread misinformation. Algorithms that amplify outrage and suppress nuance.

The response is always the same: the platform promises to fix the algorithm. Adjustments are made. The story fades. And then, a few months later, the same story appears again.

This cycle will continue indefinitely, because the people writing these stories — and the platforms responding to them — are misidentifying the problem.

The Algorithm Is Not Broken

Here is the uncomfortable truth: social media algorithms are not broken. They are working exactly as designed.

The goal of a social media algorithm is to maximise engagement — specifically, to maximise the amount of time users spend on the platform. And the algorithm is extraordinarily good at this. Billions of people spend hours every day on social media platforms, far more than they intended to.

The problem is not that the algorithm fails to achieve its goal. The problem is that the goal is wrong.

What Engagement Maximisation Actually Optimises For

When you optimise for engagement, you are optimising for emotional activation. Content that makes people feel something — outrage, fear, excitement, tribal validation — generates more engagement than content that makes people think.

This is not a hypothesis. It is well-documented. Facebook's own internal research showed that content that generated angry reactions was five times more likely to be amplified by the algorithm than content that generated other reactions. The company knew this. They chose engagement anyway.

The result is a social media landscape where:

• Inflammatory content gets amplified, thoughtful content gets buried
• Outrage is more visible than nuance
• Tribal validation is more rewarded than intellectual honesty
• The most extreme voices get the most reach

None of this is a bug. It is the predictable outcome of optimising for engagement.

The Deeper Problem: Misaligned Incentives

The algorithm problem is really an incentive problem. Social media platforms make money from advertising. Advertising revenue is proportional to time-on-platform. Time-on-platform is maximised by engagement. Engagement is maximised by emotional activation.

So the incentive structure of the entire industry points toward content that activates emotions — regardless of whether that content is true, nuanced, or good for discourse.

Changing the algorithm without changing the incentive structure will not fix the problem. The algorithm will always drift back toward engagement maximisation, because that is what the business model requires.

What a Different Incentive Structure Looks Like

A platform with a different incentive structure would optimise for something other than engagement. What might that look like?

Optimise for discourse quality. Surface content that generates substantive responses — counter-arguments, questions, evidence — rather than content that generates emotional reactions.

Optimise for intellectual honesty. Reward people who change their minds, acknowledge uncertainty, and engage with counter-arguments seriously.

Optimise for relevance. Surface content based on what people are genuinely interested in, not on what generates the most emotional reaction.

How Opiniyun Approaches This

Opiniyun does not have an engagement-maximisation algorithm. Opinions surface through categories and search — based on what people are interested in, not on what generates the most emotional reaction.

The platform's incentive structure is different from the start. Opiniyun is not an advertising platform. It does not make money from keeping you on the platform as long as possible. It is built for discourse quality — and the features reflect that.

Rebuttal Chains reward substantive counter-arguments. The Conviction Score rewards intellectual honesty. Pulse Checks give every opinion a democratic signal. The Echo system makes amplification meaningful rather than mindless.

None of these features are optimised for engagement. They are optimised for discourse.

The Alternative to the Algorithm

You do not have to accept the engagement-maximisation algorithm as the default. There are platforms being built with different design principles.

Opiniyun is one of them. Come and see what social media looks like when it is designed for discourse rather than addiction.