Want to adopt AI in your business: Good luck!
summary
Claim: 15% of AI projects are used and most failure comes from scope and ops not algorithms. Big examples: UPS saves $200M/year with delivery bundling, Google’s ad targeting drives ~$150M/year, a 2‑month POC at Pratt & Whitney identifies maintenance needs. Real build: STM ridership predictor ships in under six months by using Bombardier Azure train weight telemetry. Advice: break big visions into small projects, prioritize data quality, plan ML Ops early (roughly 80% of work is deployment and monitoring), enable continuous retraining.
AI-written summary · about
chapters (14)
- 0:00Introduction to AI adoption
- 1:06Success stories of AI at scale
- 7:50Why most AI projects fail
- 9:15Avoiding the 'Mount Everest' trap
- 12:04Breaking AI projects into steps
- 13:50Finding the right AI opportunities
- 16:57Skipping steps for business value
- 18:40Prioritizing data quality over complexity
- 22:10Change management for AI adoption
- 23:16Choosing the right path for AI
- 25:44Beyond the AI model itself
- 28:12Deploying AI in production
- 28:57AI requires continuous evolution
- 30:02Conclusion and key takeaways
related talks
37mThe AI Factor: The Blueprint for Intelligent Deployment
22mHow to operationalize AI
50mRealities & Misconceptions about Enterprise Applications of AI/ML and Data
16mHow can leaders turn operational struggles into strategic value through the integration of AI?
48mOperationalizing AI Across the Enterprise
13mAI as a Strategic Asset: Building Competitive Advantage
21mCutting through the hype. How enterprises can apply a strategic lens to leveraging AI to achieve ...
10mOperationalizing AI for business impact