๐ฅ RecovrSignal Issue 6: Clinical AI scribes just got 93% more accurate, the TGA is reviewing mental health AI, and the Support at Home clock is ticking
A clinical-grade speech model achieves 1.4% word error rate โ 93% fewer errors than general-purpose AI. The TGA opens its digital mental health review. A landmark prospective study confirms ambient scribes are ready for ambulatory deployment. And Australia's Support at Home overhaul is six weeks away.