Trustworthy AI in Healthcare and Life Sciences
summary
Claims that biased data shifts a Black patient’s liver transplant waitlist four-year waitlist shift and affects 14,000 patients, plus examples of late cardiovascular diagnosis in women, motivate concrete fixes: enterprise AI observability and guardrails rolled into Pfizer’s responsible-AI policy established three years ago; a pipeline of 200+ AI use cases; audits and monitoring tools versus the regulatory gap where only 1.9% reported safety and efficacy data on prior FDA filings.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 3:15AI bias in healthcare
- 7:33Bias in AI training data
- 11:06Need for AI governance
- 15:47AI regulations and compliance
- 21:34AI risks and emerging capabilities
- 25:52AI in pharma development
- 30:14Validation gaps in AI devices
- 35:22Bias in large language models
- 37:07Responsible AI frameworks
- 39:14Balancing AI risks and innovation