AI glossary
What is alignment?
Short answer
Making an AI system do what its developers and users actually intend, including what they didn't spell out.
For example
Asked to "get more people to click", an aligned system wouldn't resort to misleading headlines, because that isn't what its makers actually want.
Learn it in a free lesson
- Keeping AI on our side: alignment and interpretabilityAdvanced · 8 min
- Scaling laws, evals & safetyAdvanced · 7 min
Questions
What is alignment in simple terms?
Making an AI system do what its developers and users actually intend, including what they didn't spell out.
What is an example of alignment?
Asked to "get more people to click", an aligned system wouldn't resort to misleading headlines, because that isn't what its makers actually want.
Where can I learn about alignment for free?
AI in Bed explains it in the free lesson "Keeping AI on our side: alignment and interpretability" (Advanced level, 8 minutes), with no sign-up.