Mass General Brigham Deploys Hybrid Ambient AI Documentation, Cutting After-Hours Physician Work by 42%

Mass General Brigham, the largest healthcare system in Massachusetts, deployed a hybrid ambient documentation model combining generative AI with virtual human scribes across primary care. According to Mass General Brigham's published research, the deployment produced nearly 42% decreases in after-hours physician work (informally called 'pajama time') and 66% declines in documentation delays. Ambient AI documentation technologies listen to patient appointments and generate visit notes automatically. The hybrid model layers a virtual human scribe on top of the AI-generated notes for review and refinement, addressing known issues where pure-AI ambient scribes can produce errors or miss clinical context that affects coding, billing, and downstream care decisions. Before deployment, primary care physicians were completing notes after clinic hours because of documentation burden, a factor linked to rising rates of physician burnout. The team plans to extend the hybrid model to specialty care and additional sites within the integrated delivery network, while continuing to track physician satisfaction and documentation accuracy metrics. The model has implications for how large US health systems scale ambient AI safely at enterprise level, given that most prior deployments only measured pure-AI ambient scribe performance.