AI investments are expected to surge 300 percent in 2017 compared to 2016, yet most of these businesses will fail to get it right. My team recently partnered with Fortune Knowledge Group to survey 300 senior executives about their AI strategies and found that only a quarter of organizations are seeing significant impact from AI. This is more than a matter of wasted IT funds — it is an indication of a much larger issue. AI technology is a key part of business transformation, and companies that fail to get it right risk losing their competitive advantage. They are missing a huge opportunity.

The good news is that we can all learn from these initial failures. Here are some key ways to ensure your company gets the most out of its AI investments.

Prepare your workers for new roles

While AI will inevitably phase out many jobs once performed by humans, workers with the top-notch analytical skills necessary to work alongside these new machines will be more valuable than ever. Consider health care workers who have adapted to work with AI-driven diagnostic tools — they can now better avoid potentially life-threatening mistakes. Likewise, AI is creating thousands of new jobs in the energy sector for workers who can install and monitor energy-efficient devices and retrofit older buildings.

However, most workers whose jobs are set for automation are unprepared to take on these new roles, and this can be detrimental to AI initiatives. While 82 percent of leaders plan to implement AI-related technologies in the next three years, only 38 percent provide reskilling and training options to employees to prepare them for the future. Leaders need to make reskilling a top priority, as the most successful businesses of tomorrow will combine the best elements of humans and machines working together.

VB Transform 2020 Online - July 15-17. Join leading AI executives: Register for the free livestream.

To ensure that employees are AI-ready, companies must address this mismatch by investing in regular training opportunities to bring employees up to speed on these new technologies and continually adapt them to where the market is headed. For any business, developing a workforce with these analytical skillsets offers a huge edge in an AI-dominant environment. Anybody can store data, but only a few can analyze it properly and insightfully.

Identify resistance and win buy-in

Given the likelihood of robots and automation taking over many lower-level positions in the future, low and mid-level employees would seem like the most likely source of resistance when it comes to AI adoption.

Yet we found that senior executives are actually the group that most strongly resists the adoption of AI within their company. In fact, nearly one-third of our survey respondents reported senior management as the top source of resistance, compared to just 13 percent who see the most resistance from mid-level workers, and only 5 percent who view entry-level workers as the most resistant.

Ignoring the misgivings of senior leaders and managers could ultimately harpoon even the most well-thought-out AI strategy, setting a tone of hesitation and distrust for the whole company. To reduce resistance, it’s imperative leaders communicate a strong business argument for why new AI technologies will advance specific business objectives — from driving innovation to improving the quality of customer service, to ensuring safety on the factory floor. Leaders need to fully commit to AI and do so quickly if they want to future-proof their businesses.

Embrace a complete business model shift

One critical element differentiates AI success from AI failure: strategy. AI cannot be implemented piecemeal. It must be part of an organization’s overall business plan, along with aligned resources, structures, and processes.

As with any project that creates significant organizational change, having the fundamental budget and resources in place is essential. About 71 percent of companies already seeing ROI from AI-related investments are allocating the proper funds and resources to such investments. It’s impossible to overstate the importance of funding — more than one-quarter of market leaders invest at least $50 million in AI development annually, and the majority of AI leaders (57 percent) invest at least $10 million.

The reality is, AI is still relatively new and everyone is still figuring out ways to use it. However, there are certain steps companies can take to better ensure the success of their AI investments. There is no question about it — the use of AI is growing exponentially and will continue to proliferate into more areas than we can imagine today. The companies that get it right will reap unprecedented value.

Tiger Tyagarajan is the president and chief executive officer at Genpact, a global professional services firm.