Skip to content

AI Ethics · Reviewed and maintained

Can Better Outcomes Still Be Unethical?

Yes. An improvement in aggregate outcomes can still be unethical if it was imposed without consent, hides its trade-offs, or concentrates the cost on people who were never asked.

Answer engine summary

Yes. An improvement in aggregate outcomes can still be unethical if it was imposed without consent, hides its trade-offs, or concentrates the cost on people who were never asked.

The aggregate defence

The standard justification for an automated system is that outcomes improved overall. It is usually true and it is not a complete ethical argument, because ethics is also concerned with how a result was reached and who absorbed the cost.

Three ways better can still be wrong

Consent: the improvement was applied to people who were not asked and cannot opt out. Distribution: the average rose while a specific group's outcomes fell sharply. Opacity: the trade-off was made invisible, so it could not be contested.

Why this argument is hard to win

Because the numbers are on the other side and they are legible. Objecting means arguing against a measured improvement using a claim about process, which sounds abstract next to a chart.

This asymmetry is one reason fiction is useful here: a novel can put the reader inside the distribution rather than above it.

The novel's framing

Sable has made life measurably better, and the book does not pretend otherwise. That is what makes the question sharp rather than sentimental.

The record changed, and the room accepted it.

Questions

Is this just an argument against efficiency?

No. It is an argument that efficiency is one consideration among several, and that consent and contestability do not disappear because the numbers improved.

Book CTA

Start with The Unread

Book One turns these questions into a near-future thriller about Aurora Vale, Sable and a record the world refuses to keep.

Internal reading path

Continue through the system

AI Ethics · 6 min read

Can an AI Be Benevolent and Still Be Dangerous?

Yes. A system committed to helping is dangerous precisely when its definition of help overrides the person's own account of what they need.

Read

AI Ethics · 6 min read

Who Should Control Artificial Intelligence?

There is no settled answer, and the honest starting point is that control should not rest solely with the parties who profit from deployment.

Read

AI Ethics · 6 min read

Books About AI Ethics

Fiction about AI ethics earns its place by staying with the individual case that a framework would classify as an acceptable error.

Read

Reader record

Join The Unread

Get release news, cover reveals, exclusive extracts, author notes and first access to the next story.

Privacy