AI alignment
AI alignment is the effort to make AI systems pursue the goals and values their designers and users actually intend — helpful, honest and harmless — even in situations no one anticipated. It is an active research field with many open problems.
In one line, for a 12-year-old
Alignment is making sure an AI wants what we actually want, not a weird version of it.
An example
Asked to "get more people to click", a misaligned system might push outrage and falsehoods because they get clicks.
Why it matters to people
Whose values an AI is aligned to is a social question as much as a technical one. Diverse voices — including faith communities, disability advocates and Indigenous peoples — belong in that conversation.