3 big mistakes companies make with AI that limit their ROI

May 22, 2024

Create your very own Auto Publish News/Blog Site and Earn Passive Income in Just 4 Easy Steps

Opinions expressed by Entrepreneur contributors are their own.

I recently spoke with a friend who works as a CTO at a mid-sized company and was surprised by his sudden change in perspective on AI. Despite initial skepticism, he now believes that artificial intelligence (AI) will revolutionize his industry. But his biggest challenge is convincing the rest of his leadership team to adopt an AI roadmap. This scenario is not an isolated case.

Over the last year, we've seen a narrowing hype cycle around AI, leading many executives to question whether investing in AI can truly deliver proportional returns. These concerns are not unfounded. VC firm Sequoia Capital recently estimated that the AI industry spent $50 billion on Nvidia chips to train AI models last year, but only generated $3 billion in revenue.

Despite this disparity in investment, Sequoia continued to hypothesize that AI is likely “the greatest value creation opportunity” humanity has ever known, comparing its business impact to that of the cloud transition. However, unlike the cloud, which has replaced software, AI has the potential to replace services, the total market for which the VC firm estimates is in the trillions. This is why tech giants like Microsoft and Amazon continue to double down on their AI investments.

Related: What is Artificial Intelligence (AI)? Here are its benefits, uses, and more

With so many competing narratives about the future of AI, it's no wonder companies disagree about the best approach to incorporate it into their organizations. The problem is that most leaders still view AI in its limited capacity as software or a tool, rather than its ability to perform in a human-like capacity. Here are three common mistakes I see companies making when it comes to implementing an AI roadmap.

The potential of AI is underestimated and limited

AI is widely viewed as a tool or software, but because it can create and think, it has the ability to interact in human-like ways. Similar to a young employee who gets better at his job with experience, AI has the ability to learn from his interactions and refine his methods to improve his performance and work more overtime.

For this reason, leaders who consider leveraging AI through “smart people” rather than software are better positioned to realize its full potential. Think about a company's organizational chart. If you write down the skills and tasks associated with each employee, you can imagine where AI can be trained to augment or automate these tasks.

According to Stanford University's recently released AI Index report, AI is already outperforming humans in areas such as image classification, visual reasoning and even English comprehension. As of 2023, the report showed that AI has outperformed human performance on several benchmark tasks, helping workers become more productive and produce higher quality work. Another study from the University of Arkansas showed that AI performed better than humans on standardized tests of creative potential.

However, unlike humans, AI scales effortlessly with increasing business needs and handles workloads without the physical and mental limitations of humans. Adopting AI in this way means we need to rethink our team structures and workflows. This involves training teams to collaborate with AI to enhance their roles and drive innovation.

This shift in perspective is critical because it allows leaders who may not be accustomed to using technology themselves to inherently understand how to best leverage AI across the organization.

2. Attempting to emulate another company's AI use case

The more you start thinking about AI as intelligent people, the more you realize how unique each company's approach to building an AI roadmap should be. I like to think of AI implementation as onboarding new team members who need to fit into the specific dynamics of your company.

Take HR for example – a company might have 10 employees there; another only three, even if they are the same size. This difference does not only depend on the size of the company or sales. It's about how these companies have developed.

Every company has its own unique structure, culture and needs. According to PwC, to realize the full potential of generative AI, companies must leverage its ability to adapt to a company's specific needs and avoid the use case trap.

Of course, there are general use cases for AI, particularly when it comes to improving customer service or sales. However, when considering deeper integration of AI into a company's operations, the approach must be tailored and not copy-pasted from external case studies.

Related: I tested AI tools so you don't have to. Here you can find out what worked – and what didn't.

3. Buy off-the-shelf products – and don’t adapt AI solutions to your needs

There are some great off-the-shelf AI products like ChatGPT, Dalle and translation tools that solve specific problems within an organization. The challenge with investing in a boxed solution for AI is that many executives do not recognize how AI can improve operations at a systemic level.

The true power of AI lies in its ability to fundamentally change your operations, not just perform isolated tasks. According to PwC's 2024 AI Prediction Report, many companies will see attractive ROI from generative AI. Yet few will succeed in deriving transformative value from it – the biggest obstacle is the inability of leaders to think beyond ready-made solutions and reimagine the way they work with AI.

When creating an AI roadmap, leaders must first conduct a thorough assessment of their company's processes. This means identifying areas of redundancy, identifying outsourced tasks that could be automated, and figuring out where the company is investing heavily in human capital. By understanding these dynamics, leaders can adapt AI solutions to the needs of their organization and transform the way they work.

The more I talk to business leaders about integrating AI into their business, the more obvious it becomes that we as leaders need to change our perspective. By viewing AI not just as a technological upgrade, but as the incorporation of smart minds, we can better integrate it into our internal operations, increasing performance and human ingenuity.