New Benchmark Reveals AI Healthcare Agents’ Struggles with Workflows
A new benchmark shows top AI healthcare agents from OpenAI and Anthropic fail 72% of clinical workflows, raising concerns about their readiness for real-world applications.
More from this archive