1/3 GenAI projects will be scrapped

First, "there is no one-size-fits-all approach with generative AI." This cannot be emphasized enough. When integrating a generative AI project, it is crucial to recognize that this field is relatively new, and individuals with limited skills and experience in digital project integration are likely to FAIL. In my experience, every IT project I have worked on has its own unique characteristics, which distinguishes a "project" from a "product." What makes GenAI projects particularly challenging is that they represent a new paradigm. As a result, even individuals with decades of experience may struggle to effectively deliver.

Second, "AI use remains limited in scope." This limitation arises from the necessity of having solid digital experience, which includes a combination of engineering, design, and change management, along with in-depth domain expertise to effectively narrow down the processes for the use case you wish to implement. I have witnessed teams struggle with this, even when provided with ample time and resources.

Third, "determining a return on investment (ROI)" is arguably the most significant challenge in all IT projects involving modern technology. I have encountered numerous methods for measuring impact, but none have convinced me of their relevance and accuracy in assessing value or cost reduction.

There are two critical components to consider:

- the number of NEW individuals who can create new value (and were previously unable to do so)

- the number of current individuals who can be redeployed due to automation

Large Language Models represent a new paradigm; you can compare their emergence to that of mobile phones, the Internet, or even the invention of writing to understand the level of disruption they bring.

Take it the right way, and you will clearly have a leading edge. Read the full article.