Every corporate announcement seems to be about an “AI-first” launch or a new agent. On paper, it looks like a revolution. On the ground, it feels a lot more like chaos. Rushing these tools out is slowing teams down, forcing endless rework, and confusing everyone involved.
Naturally, when leaders see their teams slowing down after getting AI tools, they panic. The immediate assumption is often that the technology is broken or the strategy failed.
But moving slower isn’t a sign of AI failure. It’s what happens when you try to change how a business operates. There’s no collapse, you’re just hitting the friction phase. If you want to build a business where humans and software work together at scale, you have to expect this part and work through it.
SPEED IS A FALSE METRIC
In the rush to figure out AI, too many companies are obsessed with how fast they can roll things out and drive adoption. It is easy to see why. The pressure to move fast is real. At the same time, a culture of experimentation is critical, and that has to start at the top. As CEO, I spend time building and playing around with my own AI agents, because you can’t task your team to take risks if you aren’t doing it yourself. If your teams are afraid to experiment with new AI tools, even if it means writing bad prompts or breaking a few things along the way, you’ll be left behind before you even start.
But there is a massive difference between experimenting fast and scaling fast.
When you mistake a slick software demo for organizational change, you run straight into a wall. Rapid rollouts are causing silent chaos because executives confuse deploying a tool with people using it successfully. Moving slower at the enterprise level is required if you want to be strategic in an implementation meant to be sustained for years.
MESSY WORKFLOWS WEREN’T BUILT FOR ALGORITHMS
We talk a lot about AI-powered workflows, but if your day-to-day processes are a mess, throwing AI at them just builds a faster mess.
Our current workflows were created by humans for humans. They rely on institutional knowledge, implicit assumptions, and manual handoffs. When you drop an autonomous AI agent into that ecosystem, things break. Even if the AI does exactly what it was programmed to do, it can disrupt the entire workflow and confuse the users around it.
Plugging high-speed agents into legacy, human-centric processes naturally creates a ton of friction before you see any real benefit. As leaders, we must realize that friction doesn’t mean the tech is broken, but that the environment housing it needs to adapt.
We saw this recently with a steel manufacturing client whose estimating team was bogged down by spreadsheet-related fatigue and weeks of communication silence. They deployed an AI agent to handle pre- and post-sale operations. Technically, the AI worked perfectly; it processed email threads and task comments to generate project digests and automate weekly status reports.
But there was a learning curve. Because the team was accustomed to manual requests and sidebar conversations, the sudden shift in automation speed was jarring. The tech wasn’t failing. The human-centric process surrounding it was. The team took time to adjust, but they trusted the process. Those chaotic workflows were eventually replaced with a transparent, automated, single source of truth, allowing them to shift their focus to high-margin, solution-driven work.
DON’T MISTAKE FRICTION FOR FAILURE
The biggest risk to a company right now is a leader who mistakes this integration friction for a failed initiative.
When a new AI tool causes a temporary bottleneck or forces a process to be rewritten, impatient leaders tend to switch vendors or scrap the initiative entirely. Walking away too early kills momentum. If you constantly reset your strategy the second things get messy, your organization will never see the technology start paying off.
Surviving the AI transition requires the patience to view this messy phase as a predictable, necessary stage of transformation. Getting scale right takes time, and it requires the right building blocks. You cannot skip the infrastructure phase.
Fortunately, the friction phase is manageable. This adjustment period gets shorter when you use a platform designed from the ground up to help humans and AI work together. Instead of just throwing AI tools at disorganized spreadsheets and hoping for the best, you’re building a clear structure where human intent and machine execution align. That’s how you get to a successful outcome.
RE-ARCHITECT FOR AN AI-ENABLED FUTURE
To move past the friction into productivity, you have to stop trying to force AI into your old way of doing things. The value only comes when you redesign your workflows from scratch with AI in mind.
As you lead your team through this, also change how you measure success. Judging early milestones solely on immediate ROI misses the point. Instead, look at how well your teams are adapting. If your teams are getting better at working alongside digital agents, clearing up their data inputs, and catching mistakes early, you’re doing well. That is the exact foundation you need to scale later.
TO SUCCEED WITH AI, PLAY THE LONG GAME
The companies that flourish over the next decade will have executive stamina to sit through the messy friction phase and thoughtfully re-engineer their business for a machine-assisted world.
When you build the right foundation, where clean rules give your tools the guardrails to operate on their own, you finally fix the bottlenecks holding your team back. That’s how a smart rollout turns a system that simply tracks work into one that gets work done.
Thomas Scott is CEO of Wrike.
Leave a comment