KAIST Develops AI That Can Sense “That’s Not What I Meant” Without Being Told (IMAGE)

The Korea Advanced Institute of Science and Technology (KAIST)

KAIST Develops AI That Can Sense “That’s Not What I Meant” Without Being Told

Caption

Figure 1. Concept of Neural Value Alignment

Because human behavior can be ambiguous, conventional AI can misunderstand both the goal a person intends and the preferred way of achieving it. Neural Value Alignment leverages two types of brain responses — reward prediction error (RPE) and state prediction error (SPE) — to help AI correct both the goal and the action simultaneously.

Credit

KAIST

Usage Restrictions

No

License

Licensed content

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.