1. Focusing on AI for Its Own Sake

This is a common occurrence during every technological trend. I refer to it as "shiny object syndrome," which often accompanies fear of missing out (FOMO). There's a desire to adopt the latest and most advanced technology to keep pace with the industry, without pausing to consider its purpose.

The primary reason AI initiatives fail is that they are prioritized merely for the sake of having AI projects, rather than addressing a genuine business or customer need. When you start with a solution and work backward, it rarely proves effective compared to identifying the right tool for the problem. Leaders are prioritizing AI without considering its actual necessity.

There's also a component of peer pressure at play. If your competitors are leveraging AI, you might worry about falling behind, or shareholders might expect to see the financial benefits they’ve heard AI can deliver. While it’s not easy to disregard these pressures, the best approach is to return to the fundamental goals of your business and thoughtfully outline the best path forward.

This issue is particularly pronounced in Big Tech. We've witnessed numerous product launches, not all of which have been beneficial. These companies are often in a talent war, attempting to attract in-demand engineers, which can lead to a flurry of features that users neither need nor want.

I chuckled when I saw Cassie Kozyrkov publish her article titled “Start with Why AI,” as I am scheduled to appear on a podcast later this week discussing “The Why of Data and AI.” I couldn’t ask for a better company in this discussion!

If we fail to ask the right questions, we will not arrive at the right answers. Leaders should consider:

• What are our key business priorities?
• How does this technology provide value to the business and/or customers?
• Is this the optimal option for achieving those goals?

2. Assuming Technology Changes Everything

As previously mentioned, the first mistake is prevalent in every tech trend. Yet, leadership often treats AI as something entirely new and distinct from past technologies, disregarding valuable lessons learned during previous disruptions.

While AI and data products do have unique aspects, there's a wealth of insights from how customers, teams, and competitors have navigated past technological shifts.

Human behavior tends to repeat itself.

By reflecting on previous adoption barriers, such as those encountered with employee systems or the emergence of social media, we can apply those lessons to AI and better prepare for success.

This tendency is most observable at a macro level, where we see warnings about a "trough of disillusionment" approaching, or comparisons being drawn between the current situation and the .com bubble.

Leaders must integrate these reflections into their understanding of their organizations to effect meaningful change.

3. Viewing AI as Merely a Technological Shift

Understanding your organization necessitates understanding its people.

At the heart of any significant change is the human element.

As Sol Rashidi, author of “Your AI Survival Guide,” cautions, we must shift our focus from algorithms and big data to acknowledging that most of the work and success hinges on people and planning.

If AI represents a transformation for your business that impacts all operations, it should be treated as a change management initiative. Digital transformations frequently fail (refer to mistake two), primarily because leaders often overlook the effects on individuals and the necessity of genuine buy-in.

Employees must grasp how their roles and expectations will evolve and comprehend the intentions behind those changes.

They need to know: “What’s in it for me?”

Whether implementing AI in external products or internal processes, we must consider its impact on individuals at every stage and the behaviors that could obstruct our plans.

4. Overlooking Diverse Perspectives

Placing excessive emphasis on the views and preferences of "people like us" can lead to failure.

Especially when introducing new technology, we must be conscious of the echo chambers in which we operate.

For instance, a CEO who spends their commute listening to tech podcasts or reading AI blogs may not represent the broader customer base or their employees.

Not everyone shares the same level of awareness, interest, or trust in AI.

When Meta rolled out various AI features on Facebook and Instagram, users immediately questioned who was asking for these changes. Instead of addressing clear needs, they received confusing AI integrations that didn't solve existing problems and created new ones.

A few months ago, I conducted a LinkedIn poll about AI usage in daily jobs, where almost 90% of respondents indicated they used AI. However, my network is primarily in Data Science and Tech. Many of my family members only learned about ChatGPT recently, and most of my school peers have yet to try it.

Comments like, “What age were those who said no?” showed no clear correlation to demographics. Different functions within a company will have varying levels of exposure to new technology, and different roles may exhibit resistance to its implications. A single negative experience can drastically alter attitudes, even among those expected to embrace the change.

Do not make assumptions or rely solely on your inner circle.

Poor decisions that neglect the needs of customers or employees can lead to significant setbacks.

5. Underestimating Data Dependency for Quick Solutions

I have emphasized people-related issues intentionally, as these reflect leadership's responsibilities and highlight significant errors.

However, one final point must be addressed: data professionals also bear a responsibility to educate leadership about data's critical role.

AI cannot function effectively without data.

This is a step you cannot afford to overlook.

In the realm of open-source algorithms, data is your asset—so why would you want to bypass it?

Everyone desires rapid returns on their AI investments, but these returns may be delayed by necessary data governance and infrastructure initiatives.

We must bring the principle “Garbage in, garbage out” from data teams into the boardroom.

Leaders, if you do not allocate time and resources to a data strategy, this will inevitably become evident in your AI initiatives. I have previously written on this topic (Does your company have a data strategy?), so I won’t delve deeper here.

Google's partnership with Reddit provided a striking example of the risks involved. They spent $60 million to gather more data for their Generative AI in search, resulting in outcomes that were as bizarre as one might expect when random inputs are combined with no verification.

Leaders: Avoid being the one who pressures your team to skip the data phase.

Data practitioners: Strive to educate your leadership and business stakeholders about the associated risks. Perhaps share this article!

If you are a business leader navigating the AI hype and seeking the best path for your company, I can assist. Visit my offerings at kate-minogue.com.

With a unique focus on People, Strategy, and Data, I am available for various consulting, advisory, and fractional engagements to enhance your strategy across Business, Data, and Execution challenges. Follow me here or on LinkedIn for more insights.