EP 21 - Too much Hype, too little Impact: How to Avoid the AI Failures of Others

EP 21 - Too much Hype, too little Impact: How to Avoid the AI Failures of OthersAbout the Episode AI adoption is skyrocketing, but most AI projects don’t deliver on their promises. In this episode, Dr. Evan Shellshear, Managing Director and Group CEO of Ubidy, breaks down why 80% of AI projects fail and what organisations can do to improve their chances of success. Drawing insights from his book Why Data Science Projects Fail, Evan explores the biggest blockers to AI success, the importance of strategic alignment, and how companies can avoid wasting millions on AI initiatives that don’t deliver business impact. Whether you are an AI practitioner, business leader, or innovation manager, this episode will help you separate AI hype from reality and make smarter technology investment decisions.Evan's Books Why Data Science Projects Fail: The Harsh Realities of Implementing AI and Analytics, without the Hype: https://www.amazon.com.au/Why-Data-Science-Projects-Fail/dp/103266133X?ref_=ast_author_dp  Innovation Tools: The most successful techniques to innovate cheaply and effectively: https://www.amazon.com.au/Innovation-Tools-Successful-Techniques-Effectively/dp/0646956469?ref_=ast_author_dp Topics and Insights[02:53] – Defining Innovation Evan shares his perspective on what defines innovation, emphasising that it is not just about new ideas but the impact they create. He draws a clear distinction between invention and innovation, explaining why an idea without impact remains an invention rather than a true innovation.[04:27] – Measuring Innovation Effectively Innovation measurement starts with two key questions: Is it new? and What is its impact? Evan highlights the difficulty of measuring “newness” and discusses why impact should be measured based on purpose-driven innovation. He argues that metrics should align with a company’s innovation goals, whether it is patent creation, revenue growth, or industry transformation.[06:11] – The Biggest Blockers to Innovation in Large Organizations Evan contrasts the flexibility of small, nimble companies with the structural challenges of large enterprises. He explains how legacy systems, bureaucratic processes, and internal competition create significant barriers to innovation, making it difficult for new ideas to gain traction and survive within large corporations.[07:31] – Why AI Projects Fail: The 80% Failure Rate Evan breaks down the staggering failure rate of AI projects, estimating that up to 80% fail, with the number exceeding 90% for analytically immature organizations. He highlights the three biggest reasons AI initiatives go wrong:Lack of strategic alignment – Organizations chase AI trends without clear business objectives.Poor data quality and availability – Without the right data, even the best AI models fail.Lack of experienced resources – Many AI teams lack deep expertise, leading to unrealistic expectations and weak execution. [12:50] – The Outback AI Story: When AI Doesn’t Make Business Sense Evan shares a fictitious story from Australia’s Outback, where a farmer was pitched an AI-driven solution for managing crops. While the...