Episode Details

Back to Episodes

The 89% Graveyard: What Software's Testing Debt Can Learn from Pharma

Season 1 Episode 179 Published 2 months, 3 weeks ago
Description
Pfizer's drug development pipeline is humanity's most rigorous testing gauntlet: 10-15 years, billions in spending, and 89% failure rate. But this paranoia wasn't always the standard—it took Francis Oldham Kelsey's 1960 refusal to approve thalidomide to show why lives demand perfect safety documentation. Today we explore what software security and AI can steal from pharma's obsession with proof, and where the line between 'good enough' and overkill actually lives. 0:00 - Cold open: Entering the factory 0:30 - The drug pipeline: 10-15 years, $1.3-2.8 billion, 89% eliminated 4:00 - FDA's testing phases and regulated medicine's origin story 8:30 - Thalidomide's children and Francis Oldham Kelsey's line in the sand 13:00 - Software security's neglect vs. pharma's paranoia culture 16:00 - AI validation: disrupting or perfecting the gauntlet? 18:30 - Outro --- Sources & further reading: • FDA — "Frances Oldham Kelsey: Medical Reviewer Famous for Averting a Public Health Tragedy": https://www.fda.gov/about-fda/fda-history-exhibits/frances-oldham-kelsey-medical-reviewer-famous-averting-public-health-tragedy • Wikipedia — "2024 CrowdStrike-related IT outages": https://en.wikipedia.org/wiki/2024_CrowdStrike-related_IT_outages • IT Revolution — "Why 'Move Fast and Break Things' Is Wrong: Lessons from AWS, eBay, and Google on Speed with Safety": https://itrevolution.com/articles/why-move-fast-and-break-things-is-wrong-lessons-from-aws-ebay-and-google-on-speed-with-safety/ • Drugs.com — "How the FDA Drug Approval Process Works: Steps and Timeline": https://www.drugs.com/fda-approval-process.html • N-SIDE Life Sciences — "Average Time to Bring a Drug to Market": https://lifesciences.n-side.com/blog/what-is-the-average-time-to-bring-a-drug-to-market-in-2022 • Axis Intelligence — "AI Drug Discovery 2026: 173 Programs, FDA Framework & Market": https://axis-intelligence.com/ai-drug-discovery-2026-complete-analysis/ • Coherent Solutions — "AI in Pharma and Biotech: Market Trends 2025 and Beyond": https://www.coherentsolutions.com/insights/artificial-intelligence-in-pharmaceuticals-and-biotechnology-current-trends-and-innovations • STAT News — "FDA to Lower Number of Trials Required for Approval": https://www.statnews.com/2025/12/04/fda-considers-single-clinical-trial-for-new-product-approvals/ • SimplerQMS — "FDA 21 CFR Part 11 Questions and Answers": https://simplerqms.com/21-cfr-part-11-questions-and-answers/ • Sound of Development — "The Therac-25 Disaster: When Software Bugs Become Lethal": https://soundofdevelopment.substack.com/p/the-therac-25-disaster-when-software • CNN — "CrowdStrike Outage: What Caused It and How Much It Cost": https://www.cnn.com/2024/07/24/tech/crowdstrike-outage-cost-cause • IT Pro — "Millions of Devices Bricked and $5.4 Billion in Losses": https://www.itpro.com/software/millions-of-devices-bricked-and-dollar54-billion-in-losses-the-cost-of-the-crowdstrike-outage-continues-to-mount • CIO — "Why 'Move Fast and Break Things' Is a Liability for Critical Sectors": https://www.cio.com/article/4141933/why-move-fast-and-break-things-is-a-liability-for-critical-sectors.html • 19th News — "Frances Oldham Kelsey's Refusal to Approve Thalidomide Changed Medical History": https://19thnews.org/2025/10/frances-oldham-kelsey-history-medical-drug-safety/ • European Pharmaceutical Review — "FDA Policy Sets New Clinical Trial Requirements": https://www.europeanpharmaceuticalreview.com/news/271500/fda-policy-new-clinical-trial-requirements-us-drug-approvals/ • HBR — "What the 2024 CrowdStrike Glitch Can Teach Us About Cyber Risk": https://hbr.org/2025/01/what-the-2024-crowdstrike-glitch-can-teach-us-about-cyber-risk • Nancy Tomes (Oxford Academic) — "Frances Oldham Kelsey, the FDA, and the Battle against Thalidomide": https://academic.oup.com/book/56296 • …and 1 more in the episode research notes This podcast episode was fully generated by AI — research, script, voices, and production. Built with Claude, Piper TTS, and automated pipel
Listen Now

Love PodBriefly?

If you like Podbriefly.com, please consider donating to support the ongoing development.

Support Us