They gainsay hearing aids but there's been good progress. However sometimes it's frustrating when they "intelligently" select the wrong sound. I don't believe AI can decide better, it can be even more spurious. Unless there's eye tracker?

And there's only so much bandwidth to the damaged ear possible no matter you do. Instead I want to try glasses with sound spectral analysis/visualization to pass additional auditory information to the brain. Even with feature extraction with AI, as long as it's low latency and consistent.