Seemingly, everywhere you turn, the dominant conversation topic is generative AI. Discussion groups, webinars, conferences, email newsletters—it’s the subject everyone wants to talk about.
While the excitement and hype seem to be glaringly present in these virtual forums, it’s a sharp contrast to the level of interest and utilization we’re seeing in the real world. For example, within our own customer base, we’ve had requests from less than 1% of eligible customers to enable Microsoft Copilot.
Admittedly, we’ve been encouraging our customers to pace themselves with AI adoption, primarily to ensure they focus on reviewing and implementing security and data privacy measures before giving generative AI unrestricted access to their environment. But if our customer base is in any way a representation of the broader marketplace, then perhaps we should be thinking about why. There are a few obvious reasons that come to mind.
It’s not fully baked yet.
Tools and applications based on artificial intelligence are inherently newcomers to the market, which means they are still very much in development. We’ve seen continuous upgrades, staggered rollouts and feature enhancements across various AI-driven platforms as developers take advantage of and adjust to discovered benefits and challenges along the way.
Furthermore, many AI tools publish extensive roadmaps—meaning with all the capabilities and concerns we continue to learn about generative AI, there’s no real end in sight. That reality can make it hard to justify the cost and may be why some business leaders aren’t rushing to adopt.
It’s great for productivity, but true ROI is difficult to measure.
One benefit of generative AI that seems quite evident is that it can help users increase their daily productivity. In the case of Copilot, you can transcribe notes from a meeting you weren’t able to join or manage your inbox by summarizing email content and generating action items. Productivity hacks such as these can be helpful to the average user who spends a significant portion of their day using these tools.
But it’s also important to consider the price of these generative AI platforms and whether you’re able to earn that money back. Even if a set of employees can achieve an increased level of productivity using the provided capabilities, ensuring that this saved time is being invested back into the business is critical.
Security and data governance concerns remain.
Security and data privacy remain perhaps the most critical barrier to AI adoption. The reality is that generative AI tools, which are built into the seeds of your organization’s daily communication and collaboration efforts, cannot operate under the guise of “set it and forget it.”
A comprehensive data governance strategy is required to effectively leverage the tool’s capabilities without sacrificing security. This means implementing and maintaining strong data management safeguards such as access control permissions and content sharing restrictions.
Assessing Your Generative AI Adoption Readiness
So, is generative AI integration all hype, or is there still hope for it to be a transformative business tool? I’m still of the “wait and see” mindset and would strongly urge business and IT leaders to adopt a similar outlook, giving companies additional time to enhance these tools and address ongoing security and data privacy concerns.
For the curious customers, I advise diligent testing—preferably by only C-level or other privileged roles inside the organization. In the event that data is not classified properly, there is more limited risk of data exposure.
If you’re eager to adopt generative AI tools sooner rather than later, here are a few thoughtful questions to ask yourself first.
• What budget are you willing to spend to increase productivity without fully understanding how to measure ROI?
• How do you intend for your employees to utilize AI tools, and have you established an acceptable use policy to dictate appropriate usage?
• Have you or do you intend to train your employees on proper prompt procedure and basic AI capabilities and considerations?
• Have you completed a full data and security audit and established a data governance strategy to ensure appropriate permission structures and data privacy measures?
• Is it worth waiting another 6, 12 or 18 months to see how certain features and broader AI technology evolve?
I fully expect that AI-driven tools will find their footing and begin to demonstrate more robust and transformative applications in the near future. There’s still a lot of hope, but for now, I believe hype may be winning the battle.