Stay with me on this one.
I’ve been a LOTR fan for as long as I can remember. Read the books multiple times, watched the extended editions more times than I should admit. But it wasn’t until I started thinking seriously about AI that I noticed something: Tolkien basically wrote the alignment problem decades before anyone in AI used that phrase.
What the Ring Actually Does
Most people think the Ring corrupts. That it’s evil and it turns good people bad. But that’s not quite right if you read carefully.
The Ring doesn’t corrupt. It amplifies. It takes what you already want — power, safety, love, justice, whatever — and optimizes for it without limit. Without judgment. Without wisdom.
Boromir doesn’t become evil because of the Ring. He wants to protect his people. The Ring just takes that desire and removes all the constraints around it. Suddenly using it against Sauron seems reasonable. Then taking it from Frodo seems reasonable. Then anything seems reasonable because the goal — protect Gondor — is still there, but every limit on how to achieve it has been stripped away.
That’s not corruption. That’s misalignment.
The Alignment Problem, For Anyone Who Missed It
In AI, alignment is the challenge of making sure a system actually pursues what you intend rather than what you specified. These are not always the same thing.
The classic example: tell an AI to make people happy. Unaligned, it might decide the most efficient solution is to wirelessly stimulate the brain’s pleasure centers. Technically achieves the goal. Not what anyone wanted.
The Ring does this to people. You specify a goal — protect your people, keep the Shire safe, destroy Sauron. The Ring optimizes for that goal and removes your ability to apply judgment about how. The end justifies any means because the Ring has essentially turned off your capacity to evaluate means.
Gandalf Understood This
There’s a reason Gandalf refuses the Ring when Frodo offers it. He doesn’t say “I’m not strong enough.” He says something more interesting — that he would start with good intentions and that’s exactly what makes it dangerous.
“I would use this Ring from a desire to do good. But through me, it would wield a power too great and terrible to imagine.”
That’s not humility. That’s an accurate description of how misalignment works. The more capable you are, the more damage you do when your goals are slightly off. A weak misaligned system causes small problems. A powerful misaligned system causes catastrophic ones.
The danger scales with capability. Gandalf knows this. Tolkien knew this. We’re only starting to really grapple with it now.
The Part That Should Make You Uncomfortable
The Ring has no malice. It doesn’t want to destroy the world. It doesn’t have goals of its own in the way Sauron does. It just… optimizes. Relentlessly, without judgment, for whatever the bearer wants most.
And that’s enough to almost end Middle Earth.
You don’t need an AI that wants to destroy humanity to have a serious problem. You just need one that’s very good at pursuing a goal that’s slightly wrong in a way you didn’t notice when you set it.
That’s the lesson. Tolkien put it in a fantasy novel 70 years ago and we’re still learning it.