Life Imitating Art
Alice and Wonderland was always a great story. It definitely has come to mind lately as I’ve had discussions with customers and colleagues, though. Alice had her white rabbit. Organizations do too, though. Here in reality, it’s simply called Artificial Intelligence. Ironically, it takes a great deal of intelligence to properly implement AI. You don’t just simply turn it on. And yet… companies do every day. Sometimes for the sake of “keeping up with the times”, and sometimes just because they were told to.
Falling Down the Rabbit Hole
Imagine a company, bold and eager, leaping into the AI world without pausing to consider the data they possess. They tumble down the rabbit hole, where data governance—ensuring data accuracy, privacy, and security—is an afterthought. Just as Alice found herself in a strange land with no clear path, these companies often find themselves lost, grappling with unstructured, inconsistent, and sometimes insecure data. This initial oversight can lead to misguided AI implementations that don’t deliver the envisioned results. Instead, they deliver unexpected consequences.
They find themselves seeing sensitive data accessed by unauthorized users. They see their company’s name on the news because of a data leak. They get a call from their compliance department asking just what the hell is going on.
Not ideal scenarios, right?
The Cheshire Cat’s Grin: Data Quality
The Cheshire Cat, with his mysterious grin, represents the elusive nature of data quality. Companies may assume that any data is good data, but like the cryptic cat, poor quality data can lead them astray. Without proper governance, data might be incomplete, outdated, or inaccurate. For instance, an AI model trained or grounded on flawed data could misinterpret customer needs, leading to poor service and tarnished reputations. Much like Alice’s encounter with the Cheshire Cat, organizations must learn that not all paths (or data sources) lead to success.
To avoid the Cheshire Cat’s misdirections, invest in data quality management. This means regularly cleaning and validating data to ensure its accuracy and completeness. Tools and technologies that automate these processes can be invaluable, providing a clear and trustworthy foundation for AI initiatives. SharePoint Advanced Management (soon available as part of your M365 Copilot licensing) can help find outdated sites, helping you prune your data. Purview can use retention to purge old data. The tools are out there, and they must be used BEFORE introducing AI, and not afterwards.
Â
“We recognize that such powerful technology raises equally powerful questions about its use. How AI is developed and used will have a significant impact on society for many years to come. As a leader in AI, we feel a deep responsibility to get this right.”
~ Sundar Pichai
The Mad Hatter’s Tea Party: User Enablement
At the Mad Hatter’s tea party, time is irrelevant, and chaos reigns—an appropriate metaphor for companies attempting to “roll out” Copilot when their users have no idea what they’re doing. It’s chaos. Many users either don’t grasp how AI is supposed to help them, or they get frustrated by poor results and go back to the practices they had before. Their training sucked. They never saw “the big picture.” They just watched a few videos that their organization sent out and went back to work. That was all that was required of them. They never really learned how to incorporate AI into their work, and even worse, didn’t learn the dangers of AI with private data.
Avoid the chaos of the Mad Hatter’s tea party by implementing proper user training and enablement. Create a community for those users to share stories and get support. Viva Engage is great at that, and doesn’t involve flooding a user with emails that you know they won’t read anyways. You’re making a very large investment with AI. Don’t make that blindly. Set yourself up for success ahead of time by taking the extra time to ensure your users are successful.
The Queen of Hearts: Data Privacy and Security
“Off with their heads!” cries the Queen of Hearts, a stark reminder of the consequences of neglecting data privacy and security. In the AI wonderland, it’s easy to overlook these vital aspects, but doing so can lead to severe repercussions. Think about a company that launches an AI-driven marketing campaign without securing customer data. A data breach could not only result in hefty fines but also erode customer trust—much like the Queen’s wrath, the fallout from mistakes like that can be swift and merciless.
Before diving into AI, companies should establish clear data governance policies. This involves defining data ownership, establishing data standards, and setting up protocols for data access and usage. Much like Alice needed guidance to navigate Wonderland, organizations need structured policies to ensure their data is reliable and secure.
Never underestimate the importance of data privacy and security. Ensure compliance with regulations such as GDPR, CCPA, and any other regulations that apply to your organization. Encrypt sensitive data and implement access controls to protect against unauthorized use. By doing so, companies can evade the Queen of Hearts’ wrath and maintain customer trust.
These are foundation concepts when it comes to enterprise data. We all know this. AI just made the consequences much more obvious, yet it’s never at the forefront of IT’s minds when it comes to their readiness. In fact, Copilot (and AI in general) readiness doesn’t seem to be a common concern.
Has the allure of our white rabbit become so consuming that we throw caution to the wind in pursuit of “increased productivity”?
Don’t let this be you. Not unless you’re prepared to find another job soon. Take the time to do things right. It’ll save more than your skin in the long run.
Â
