The news cycle has been abuzz lately about Artificial Intelligence (AI). First there was ChatGPT, then BingGPT, then Bard from Google. Now, Microsoft has announced Copilot for Office. And I suddenly feel like AI just got real.

With ChatGPT and its ilk, we have been treated to examples of AI-generated stories and media. I agree that these are interesting for the future they portend. However, I do not write for a living (insert snarky joke here). I want AI to help me with more common tasks. This is where Copilot for Office sounds interesting.

Copilot Aims to Marry AI and Everyday Tasks

Allow me to pump the brakes for a moment. Copilot is yet to be delivered. So, we are seeing examples of what Copilot promises to do. However, we have not seen actual results yet. Still, the promise looks interesting. Here are some examples from a Copilot video Microsoft released.

• Copilot processes the audio and video inputs from your Teams meeting. It produces a set of meeting notes that includes key images, action items assigned to participants, and a project plan for next steps.
• Copilot takes my draft security assessment report (100+ pages) and generates an Executive Summary.
• I feed the same security assessment report into Copilot. It uses the information to generate a PowerPoint presentation for executive management.
• Copilot reads my emails and generates a summary of key messages for me.
• I ask Copilot to look at my Excel data and tell me what charts or pivot tables I should create, to better understand the data.

If Copilot can live up to its promises, I will be impressed.

Watch This Space

Whenever I see these breathless announcements about one or another “disruptive” technology, I think back to Amara’s Law:

We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.

This same thought is what underlies the Gartner “Hype Cycle.” Early on, the technology seems to show that anything is possible. As we dig in and start using the technology, we begin to discover its limits. Eventually, we calibrate our reactions and expectations.

This round of AI announcements, for all their hype, does feel different. We are starting to see AI do things that go beyond narrowly defined use cases. And the pace of innovation is rapid. ChatGPT is already based on an updated version of the AI engine. Microsoft moved quickly to incorporate ChatGPT into their Bing search engine. And the incorporation of Copilot into GitHub is already helping coders with suggestions for new code.

I will be interested to see what I can accomplish with Copilot once it is released. Will it change the way I work? Or will it offer me the skills of a below-average copy writer? We will see.

I hope you will try Copilot when it is widely available. See if Copilot can take on some routine tasks. Even better, see if it can help deliver some key outcomes using a redefined process and some automation.

Peel back the hype and take Copilot for a test drive. You may like what you find.