As we enter into the 2nd half of the year, companies are more confused than ever about what AI is and what it actually does. Artificial Intelligence has been around in some form or another for almost 70+ years, and if you really want to get technical, its history goes back to the beginning of time. Everything that has been invented and lasts uses machine learning in some form or another. This may seem abstract to a degree, but how many times did the steam engine get built and rebuilt until it was ready to be used widely? To perfect all of its parts and ensure the engine worked took thousands of hours to plan, build, and optimize. AI is no different. Instead of parts to make up the engine, AI needs data to build its large language models to provide predictability and reliability.
The hype cycle around AI, particularly Generative AI is at an all-time high, and with good reason. GPT-4o was announced in May, and the pace of change I am seeing in the software development space is astounding. New paradigms and applications are ready to be built. Critical decisions are being promised, and the world is changing under our feet.
The question I ask is, are we even ready for this, and do people even understand what “AI” really is?
It’s a fair statement to make. Most times when I hear our customers talk about AI, they are really talking about something else. They can say I want AI to automate said task, which requires business rules and directions to have it do the job on a desktop. That is what we know as RPA (Robotic Process Automation). I hate to say it, but our old friend RPA is here to stay for the time being, until every application known to man can be automated from within, and fully capable APIs are readily available. If you take a peek at any enterprise company, 70% of their IT spend is devoted to maintaining legacy applications. That may change over time, but it won’t happen soon. Mainframe computers are still powering finance, healthcare, the power grid, and the public sector to name a few.
I think for this reason, companies need to be realistic about their short-, medium- and long-term goals. As stated before, most companies haven’t started their Intelligent Automation journey. Many have struggled to get it off the ground because of two major factors: bad choices of processes to automate and a lack of a support strategy. Both continue to be reasons why AI/RPA will continue to get a bad rap. The need for intelligent applications that power streamlined workflows is widespread, and how we get there will be a journey that many forms of AI, RPA, data, and cloud-based architectures can all help with.
Written by: Peter Camp, CTO & Founder
