Rapid growth, large investments, high excitement, and even higher expectations. These last few years have been ripe with these feelings and actions, in regards to AI. Now that it’s long entered the mainstream, and seemingly everyone is using it—or trying it out—people are starting to ask: 

Are we in an AI bubble? Will it burst? And how will that potential burst impact me? 

In our current AI-fascinated/AI-disillusioned world, 79% of corporate strategists believe that technology like artificial intelligence would be critical to their success in the next two years. However, widespread economic impact is still a few years away. Goldman Sachs suggests that AI adoption will likely begin its meaningful impact on the U.S. economy as a whole, between 2025 and 2030.

Additionally, recent publications from the Goldman Sachs report “Gen AI: Too Much Spend, Too Little Benefit?” to TechBrew are starting to wonder if we’re in an AI Bubble. They’re starting to ask questions surrounding the validity and payoff of AI right now . . . and they should!

Here, examine exactly where we’re at in the AI hype cycle. And beyond the hype, we’re discussing 3 action items for all businesses to incorporate into their AI vendor selection process . . . whether we’re in a bubble or not!

The AI Innovation Cycle AKA The AI Hype Cycle

The conversation surrounding AI has shifted dramatically over the last several years. Remember when ChatGPT and Dall-E were absolutely mind-blowing? Well, they still are, but they’ve become part of our everyday lives. We’ve since moved out of AI’s Early Hype phase, which was full of excitement and optimism that drove rapid investment and development.

We’ve also passed the Peak of Expectations, which was marked by both extreme enthusiasm and the initial discussions full of anxiety surrounding AI’s potential negative impacts.

Today’s conversation surrounding an AI Bubble shows that the general public is entering the Trough of Disillusionment with AI. Here, ambivalence becomes more prominent. Initial excitement has waned and challenges surrounding the technology are becoming apparent.

In the future, enthusiasm will rise again (in the Slope of Enlightenment phase), as a more balanced emotional approach to AI will emerge, with realistic assessments of AI's potential and limitations. And, finally, emotions stabilize as AI becomes more integrated into daily life in the Plateau of Productivity. 

Long-Term Economic Impact is Great. But You Can’t Wait Around For It. 

The truth is, while public sentiment shifts and investors’ and business’ purse strings tighten, AI solutions will continue advancing. Coupled with this continuous innovation, experts still agree that AI’s economic impact will be seen in big ways. 

For example, PwC's Global Artificial Intelligence Study study estimates that AI could contribute up to $15.7 trillion to the global economy by 2030. This contribution is expected to come from increased productivity ($6.6 trillion) and consumption-side effects ($9.1 trillion). Their report specifically emphasizes AI's role as a key source of competitive advantage in the economy.

Yes, AI will pay off in the long term. But of course, you need to see a payoff from your investment in it now.

The good news is that you don’t need to wait around for the next phase in the AI Hype Cycle in order to experience a positive financial impact from the technology. You do, however, need to be highly selective of the AI solution to trust and partner with. After all, despite significant advancements in technology, AI still faces many technical challenges and limitations. Over-promising and under-delivering has led us to this current state of disillusionment.

Here, we provide three action items to keep in mind when selecting an AI partner. You can apply these to your process now or wherever the AI Hype Cycle takes us!

Action Item #1: Select a partner with sustainable, achievable goals. 

The truth of the matter is that you need to be very careful on what AI solution you partner with. You simply can’t risk working with an AI solution provider that promises too much, too quickly. 

In this upcoming phase of the AI Hype Cycle, the not-so-great AI solutions will have their downfalls. In turn, this means that the true leaders in AI will survive, and not only that, but they will thrive. If we’re in an AI Bubble, you can compare it to the 1990s Dot-Com Bubble. At that time, many companies went belly-up. On the flip-side though, sustainable, truly impactful companies—Amazon, eBay, and Priceline, to name a few—continued to grow. 

If we have another tech bubble burst, the players in the AI space with sustainable business models, measured growth plans, and provable impact will follow in the Dot-Com survivors’ footsteps. 

Action Item #2: All AI solutions need to prove measurable impact when solving for your specific needs.

As you evaluate potential AI vendors, it's imperative to learn about their sustainable plans for long-term growth. We’ve moved past the days of AI being a shiny object and board mandate. It now needs to prove its worth. 

As you set out on your AI journey, define specific business goals and outcomes you need to achieve with AI. The Computing Technology Industry Association’s AI Advisory Council says that, "defining milestones for an AI project up front will help you determine the level of completion or maturity in your AI implementation journey." As you define these needs, communicate them clearly with your potential AI vendor while evaluating if they can help you achieve them. 

Importantly, look for AI solution providers with proven experience in—and knowledge of—your industry. Their understanding of your industry-specific challenges are critical, as this shows that they truly understand your functions and pain points. We suggest you ask about case studies and real, measurable success with their current clients.

Action Item #3: When it comes to data and ethics, make sure your AI partner is the real deal. 

Even beyond understanding your needs, pain points, and industry, the AI partner you’re interested in has to have the right data powering their tools. When evaluating vendors, ensure that their solutions’ models are fine-tuned on your industry’s data, so they can therefore provide specific outputs that are helpful in solving your needs.

We suggest that, when meeting with sales teams, ask if you can speak with a technical expert in engineering or data science. Inquire about the following:

• Where does their data come from? How frequently is it groomed? How is it maintained?
• How will your data be used, especially if it might be incorporated into the AI provider's training data?
• How does their AI system make decisions and mitigate potential biases?

Truly helpful vendors will be transparent regarding the technical specs of their solutions. This trustworthiness sets realistic expectations for your own AI-powered growth plans, which is absolutely critical to have, whether we’re truly in an AI Bubble or not!