When AI joined the IT department
When I first heard about AI, I was skeptical. It felt like a gimmicky Siri or Alexa. Over time, it’s become something else entirely, a daily tool that I now can’t imagine working without. In fact, AI might be the closest thing we have to a second brain in IT, if (big “IF”) you know how to use it properly.
I’ve spent two decades in IT across small and medium-sized companies, and I’ve seen plenty of shiny tools come and go. AI is different. It’s not just a tool, it’s a level-up enhancement. But like the old metaphor, when you’re a hammer, everything looks like a nail! You’ve got to develop a keen sense about how and when to use it.
Institutional knowledge and emotional intelligence will continue to be the superpowers for us humans.That’s especially true in IT, where our human judgment and experiences are critical, especially with the users and businesses we support and, over time, grow to understand. It’s these fine nuances that AI won’t be able to navigate. But here’s how you can find ways to work with AI for the good of you and your company.
AI is not the answer, it’s a tool
Let’s be clear: AI is not “the answer.” It’s not going to replace your years of experience, gut instincts, or the long hours spent figuring out why a random application kept crashing. What it will do is get you 70% of the way in 10% of the time. That remaining 30%, that’s all you! That’s what makes us in IT truly invaluable.
Around April of 2023 I was attempting to recreate templates and documentation to stuff I lost like training outlines, and internal policies that I had fine-tuned over the years. I didn’t want to dig through my various digital junkyards or rebuild it all from scratch. So, I dumped everything I remembered into ChatGPT in a super long run-on sentence, with no real structure or organization, and asked it to organize, structure, and polish.
What came back was certainly not perfect, but it gave me a strong base I could tweak and build from to make my own again. It reminded me of ideas I hadn’t used in years. It made a structure of my babbling and random notes. I started to compare some of my old output documents side-by-side with the new one and added in additional specific business and user insights.
The result was a set of documents that improved on my originals, but they still maintained my style and expertise, along with addressing the needs of the task. Ultimately, this led to a process that saved me time and gave me a skill set that I still use to some degree today.
As new and exciting as this was, I still circled back to one thing: AI could not have done this without my original input. The documents were still the product of my expertise, just made better with the help of AI.
Accuracy > speed: Don’t fall for the echo chamber
One of the worst, and usually first, things I see AI novices do is take the original answer to a query as gospel truth. AI is coded to always give an answer. AI is confident, even when it’s wrong. Some AI users don’t know about hallucinations or don’t understand that as you work through a longer issue with AI, the conversation drifts. This is another example when your experience is a necessary ingredient for AI to contribute. In the end, you need to validate the advice it gives you with your own experience, or that of your colleagues.
Use it, own it, stay accountable
AI won’t replace IT, but I’m optimistic about its ability to elevate myself, my team, and the businesses I support. Accountability must still rest with a human who has the experience and the knowledge to implement AI correctly while understanding the needs of the business. We are still the break-fix experts. We must still be the ones who use our emotional intelligence and our business acumen to explain issues to the C-level or the end-user in a language that resonates. We can’t offload responsibility to AI when we need to answer for our systems and issues with our processes.
Embrace AI. Learn and study it. Stay in the loop. If you use it correctly, AI can help you become more productive, more accurate, and more confident, without replacing you and your experience. We are that firewall between rushed AI output and a well-executed decision. We still take ownership of the processes or documentation for the businesses we support. In today’s IT world, that’s what leadership does.
