When it comes to AI adoption, go slow to go fast

The artificial intelligence (AI) revolution is here to stay—42% of companies are exploring AI adoption and many have already started implementing it. Organizations will want to rush to be the first to implement AI, but they must focus on doing it effectively and sustainably. As AI rapidly advances, imagine if large language models—AI tools that use vast amounts of data to generate new content—continue to progress at a rate comparable to iPhones. An organization could end up purchasing or adopting a new version every year, compounding costs and causing frivolous disruption not anchored in a clear understanding of what that investment is contributing to the organization. 

Instead, leaders should focus on a design thinking approach to embrace AI transformation incrementally and organically, so it is built upon a strong and appropriately contextualized foundation for future resilience. 

The Navy SEALs have a valuable motto: “Slow is smooth, smooth is fast.” This principle underscores the importance of precision in achieving objectives and can be applied to the planning process for technology-driven operational changes. Our data and more than 20 years of experience partnering with organizations to drive enterprise-wide, transformative change indicate that leaders should adopt AI incrementally rather than attempting a wholesale transformation. By taking measured, incremental steps, organizations can achieve their goals more effectively and efficiently, and ensure that employees are given the proper resources to successfully integrate the technology into their roles.

Short-term decisions, long-term consequences

There’s no question that AI can be a gamechanger for organizations by streamlining processes and highlighting insights and patterns that are difficult for humans to see. But many are getting swept up in the shiny object appeal of AI adoption, and they commit to certain tools quickly. 

Constant updates and abrupt implementation without properly acclimating employees can lead to worker burnout, resentment, and turnover. If AI isn’t properly introduced or deployed in the right areas, staff may be hesitant to adapt and give their full buy-in, which is vital to successful implementation. 

Speeding to full-scale AI or automation can also present risks in safeguarding human-centered process controls, without which organizations are more prone to face worrisome mishaps that can mar a brand reputation and employee morale in the long term.  

Consider the recent example of Dukaan CEO Suumit Shah firing 90% of his support staff in favor of a chatbot, and broadcasting the news on Twitter that the average time taken to resolve a customer’s issue dropped by almost 98%. While he described the decision as purely financial, Shah faced swift backlash after countless users tweeted that his support team losing their income stream was nothing to celebrate. 

Or consider Vanderbilt University using ChatGPT to pen a statement on the Michigan State University shooting. The letter mentioned creating a culture of respect and understanding on campus, and ended with a sentence in parentheses that read, “Paraphrase from OpenAI’s ChatGPT AI language model, personal communication, February 15, 2023.” Critics countered that a statement on such a sensitive topic should be grounded in human empathy and respect.  

The lesson from these stories is that short-term decisions can have long-lasting, unintended consequences. While AI and large language models are surely going to help organizations improve, it’s critical that they adopt AI in a thoughtful, human-centered way that leads with empathy throughout AI implementation.

Organic AI adoption beats an AI overhaul

Quickly implementing AI can lead to wasting resources on tools that don’t actually help your workforce accomplish organizational goals. Organic integration, on the other hand, can boost performance by incorporating AI tools that actually improve productivity without harming your organization’s reputation or morale. And when AI is integrated humanely, positive employee experience is preserved.

Starting small might look like introducing training to use out-of-the-box large language models (LLMs), such as ChatGPT, as a desktop job aid. Or implementing robotic process automation (RPA) to regulate data quality and manual work activities before embracing large-scale AI adoption. RPA is an automation technology that can take the work people generally dislike off their plates, for instance, logging into applications and scraping data from the web. RPA makes people more productive by taking care of mundane work, while still leaving important decision-making processes that require human empathy to the individual. 

By automating discrete pieces of the process and monitoring the results, leaders can have a clearer picture of where AI works well and improves decision making.

Staying thoughtful

AI adoption is upon us; it’s important that we integrate it into our organizations thoughtfully and prioritize people along the way. The litany of AI mishaps in the news are no coincidence. As AI adoption ramps up, countless companies will rush to adopt it without stopping to think about long-term effects. The most successful organizations will think about how to implement AI in a thoughtful, measured, and human-centered way that prioritizes their employees.