Cautious Approach for AI in Aviation

The chairman of the House Artificial Intelligence Task Force, Rep. Jay Obernolte (R-California), recently addressed the audience at Honeywell’s American Aviation Leadership Summit in Washington, D.C., asserting the potential usefulness of AI in aviation. However, he urged caution, emphasizing that trusting AI is a different matter entirely. Some industry stakeholders believe AI implementation is imminent, while others estimate it may still be over a decade away.

Obernolte highlighted the importance of ensuring that AI deployment is both safe and intelligent. “We need to do everything we can to put ourselves in a place where we can keep pace and deploy AI safely and intelligently,” he stated.

During the session, David Dunning, director of global innovation and policy at the General Aviation Manufacturers Association, posed a critical question: should AI face challenges? “Sure,” Obernolte replied, underscoring the necessity of skepticism regarding AI accuracy. It is crucial to never, ever assume that AI is correct, particularly due to the broad and fallible nature of the information on which it is trained.

The House AI Task Force advocates for regulatory oversight of AI in fields like aviation through existing agencies such as the Federal Aviation Administration (FAA). Obernolte insisted that industries must establish guardrails to mitigate risks associated with AI use.

Obernolte, who holds a master’s degree in AI from UCLA and is both a Certified Flight Instructor (CFI) and a commercial helicopter pilot, pointed out the importance of careful training for AI systems. He emphasized that Scale AI, often overlooked, plays a crucial role in providing the right data necessary for training algorithms effectively.