As organizations race to adopt AI, many leaders are asking the wrong question. Too many are using AI to optimize yesterday’s workflows instead of asking a harder question: should this process still exist? Using AI in small, incremental ways—like summarizing emails or automating tasks—may create some efficiency, but it rarely creates meaningful impact.
True value doesn’t come from layering AI onto existing processes. And that distinction matters as 60% of companies report little value from their AI initiatives, despite significant investment.
Organizations should approach AI less as a technology deployment and more as an opportunity to rethink how work gets done from the ground up. Instead of asking how AI can improve a process, leaders should ask: If we were building this workflow today, with AI available from the start, would we design it the same way?
In many cases, the answer is no.
Many AI initiatives start with efficiency, and those improvements may ultimately save time, but they rarely change the outcome. The bigger opportunity is to expand what organizations believe is possible with AI.
I see this with my own team. In the past, we may have had the capacity to prototype two or three ideas at once. Today, AI can dramatically increase that capacity and change the scale of what we test and build. That doesn’t mean we shrink the team; it means the expectation for output and impact evolves.
That shift, however, requires new collaboration. The people closest to the work understand the nuance, exceptions, handoffs, and outcomes that matter. But they also live inside existing processes every day, which can make it difficult to see beyond the current workflows or immediate impact of small improvements. They must be paired with teams who understand that AI creates an opportunity to challenge assumptions, remove outdated [constraints], and redesign work around better outcomes.
Working together, the group can ask these questions to test their approach:
A useful litmus test is whether the new process materially changes the outcome. If the only benefit is that the task takes less time or requires fewer clicks, the organization may simply be adding AI to an old workflow.
On my own team, AI has dramatically expanded our ability to test ideas. What once might have been two or three concepts can now become dozens of prototypes, allowing us to challenge long-held assumptions about how work should happen.
That experimentation creates the opportunity to do more than improve an existing process. One of those ideas evolved into Cengage’s Instructor Assistant, not simply as another AI feature, but as a way to rethink the instructor workflow itself. The goal was never to just give educators another dashboard or more insight into student performance. It was to help instructors understand what action to take next. By turning student interactions and assessment results into contextual recommendations, instructors can identify learners who are struggling earlier and intervene more effectively. The value isn’t the insight itself; it’s what the instructor can now do because of it.
The same principles can be applied across industries.
In customer service, for example, layering AI onto an existing workflow might mean using it to summarize a call. Redesigning the workflow means using AI to understand the issue, gather customer [context], and then route the call to the right agent from the start. Cisco is a great example of this shift, bringing AI, insights, and support together across the customer journey rather than simply accelerating individual tasks. The outcome is not just faster support, but better customer experience overall.
The same is true for enterprise operations. Organizations can move beyond automating tasks and instead use AI to identify patterns, surface risks earlier, and help leaders make more informed decisions. IBM’s recently introduced AI Operating Model reflects this approach. The outcome is better visibility and more thoughtful use of people’s time.
The organizations that create the most value from AI will not be the ones with the most pilots, models, or tools; they will be the ones willing to rethink how work gets done.
As AI becomes more accessible, technology alone will no longer be a differentiator. The advantage and impact will come from redesigning workflows and decision-making around meaningful outcomes.
AI may accelerate the work. But reimagining the work is what creates transformation.
Darren Person is chief digital officer at Cengage.