National Institute of Standards and Technology Plans Additional AI Evaluation Measures

Elham Tabassi, the National Institute of Standards and Technology’s associate director for emerging technology, told lawmakers that she is planning to introduce new assessment techniques and a standards tracker for artificial intelligence.

Her efforts are part of NIST’s goal to promote the development of responsible AI, Nextgov/FCW reported.

Testifying before the House Science, Space and Technology Committee on Wednesday, Tabassi explained that her office is studying AI systems’ socio-technical aspects to determine if they can be deployed safely. Upcoming metrics will enable developers to determine the level of harm their systems could cause, she added.

According to Tabassi, the new standards tracker will support “NIST’s existing matrix” to provide a common foundation for AI systems without broad regulations. She did not specify what data the tracker, which will be part of the AI Resource Center, will collect to monitor standards.

Tabassi said her team’s efforts are meant to help operationalize the AI Risk Management Framework, especially for small and medium-sized businesses.