Ep161: Why 5% of AI Projects Succeed - And How Agentic AI Changes the Game

Qlik's Field CTO for Generative AI Ryan Welsh reveals why 95% of enterprise AI projects fail and shares the three proven strategies the successful 5% use to deliver real business value from their AI investments.

Topics Include:

• Qlik's Field CTO reveals why 95% of AI projects fail despite massive investments
• MIT research shows shocking failure rates, but 5% are achieving real business value
• First major pitfall: Bad data foundations doom even the most sophisticated AI models
• Second problem: Companies use generative AI when predictive models would work better
• Third issue: Unnecessary complexity - AI projects disconnected from business outcomes
• Success secret #1: Ground AI in trusted enterprise data and user context
• Some LLMs struggle at specific tasks like claims processing despite passing medical exams
• Success secret #2: Let AI learn from users while keeping data governance intact
• Success secret #3: Embed AI directly into existing workflows like Salesforce
• Agentic AI shifts from reactive Q&A to proactive systems that execute across platforms
• Case study: Lintek reduced churn 10% and saved millions using these principles
• Your AI choices today will lock in your trajectory for years to come

Participants:

• Ryan Welsh – Field CTO – Generative AI, Qlik

See how Amazon Web Services gives you the freedom to migrate, innovate, and scale your software company at aws.amazon.com/isv/