Definition
The research field focused on ensuring highly capable AI systems operate safely and reliably, adhering precisely to human values and intentions.
In short: The research field focused on ensuring highly capable AI systems operate safely and reliably, adhering precisely to human values and intentions.