COMMON POWERUp 2026

More and more companies get into trouble with their old applications and this is not only on IBM i. The reasons are different but at the end they all have the same challenges: They need to get out of that. One way of doing that is using Tools to analyze the old stuff. That is something we all do for decades now but the tools changes or continue to evolve. Right now there are many "Tools" out there and they all claim to use AI and many of them are based purely on AI. All of them promise you, that they will solve all your problems. My team and I have spent a lot of time and energy over the last two years looking at many of these tools (and those that claim to be such) and have done a lot of experimenting and developing in this area ourselves. Many of these tools deliver very questionable results, which sometimes fit perfectly and sometimes don't fit at all. However, our customers and we need to be able to trust the results of such a tool. Is it true or false? Not sometimes yes and sometimes no... In addition, many of these tools incur enormous additional costs, especially if they rely on cloud-based AI models, where you can never be sure what will happen to your company data. I can show you a working solution that is used in production for many months and can use pure local AI Models combined with "self made" development to create results we can trust.

Learning Objectives:

See why it is important to take a close look at how such tools work, understand them, and be able to compare them, as the results and costs of the various tools vary enormously. You need to be able to recognize the advantages and disadvantages of the various tools and decide which one is best suited to your situation and your company. This is not only a question of security, compliance, and valid results, but also of TCO. I will show you examples of some of the above points live so that you can see exactly what I mean. See how you can analyze your old and very old code with the help of a tool that not only uses AI, but only where it makes sense. And it works now—not sometime in the future.