The AI Conversation Is Stuck On the Wrong Question. Here’s the One That Actually Matters
There are two ways to use AI inside a team. The first is the one most organizations default to: hand AI the tasks people already do—drafting, research, summarization—and collect the efficiency gains. It works, but there’s a ceiling.
The second is more strategic. It means stepping back from the tasks entirely and asking a different question: now that AI can handle parts of this process, does this workflow still make sense? This question is about looking at the whole system and deciding which steps should exist, how they connect, and where AI or automation should carry the weight.
The teams pulling ahead right now have learned to think at that level. But that capability rarely emerges on its own. Leaders who recognize it, develop it, and build it into how the team operates are the ones who turn individual insight into organizational advantage.
Using AI and Rethinking the Work Are Two Very Different Things
Leaders need to distinguish between two modes of progress: using AI within existing work and rethinking how that work happens in the first place because of AI. Both matter, but they operate at different levels.
(Shutterstock AI)
- Using AI (task-level) focuses on accelerating discrete steps (think: drafting, summarizing, and research) within existing workflows. It requires little structural change and delivers immediate efficiency gains. The impact shows up as speed, mostly at the individual level.
- Rethinking the work (workflow-level) centers on identifying what’s possible with AI that wasn’t before. It involves reconsidering which steps need to exist, what order they should happen in, and where a person actually needs to be involved. The impact shows up as capacity and new opportunity—teams can take on work they previously couldn’t, serve needs they didn’t have bandwidth for, and operate at a scale that wasn’t realistic before.
Teams that stay focused on tasks see an early boost in productivity. But those gains plateau. Bottlenecks persist, handoffs stay manual, and gains flatten out. Over time, keeping up demands more effort, including additional tools and incremental fixes, without changing how the work actually operates.
When you build the skill to rethink the workflow itself, you increase your capacity to take on more complex tasks at scale. The same group can produce more, with fewer constraints, because the system is doing more of the work for them.
How Leaders Can Spot, Develop, and Scale This Capability
Every team already has people who instinctively rethink how work gets done. They’re often hidden in plain sight: quietly compressing timelines, stitching together tools, and building lightweight systems that make their output look effortless.
The opportunity for leaders is to make that way of working visible, repeatable, and scalable across the team. Here’s how.
1. Evaluate how work gets done (not what gets done)
Most performance conversations start with outcomes. The process comes up occasionally, but it’s rarely examined in detail. As a result, meaningful differences in how work gets done—manual versus automated, fragmented versus streamlined—often go underexplored.
Start making workflows part of how you evaluate work. In team reviews, ask: What did this replace? Which steps are no longer needed? What runs without direct involvement?
This makes work easier to analyze and improve because the underlying system is visible. Instead of reinforcing task execution, you create space for redesign.
2. Build “workflow literacy” across the team
People need to understand how to break work into components, where automation fits, and how AI can handle parts of the process reliably. Without that baseline, even the best tools remain underused.
(Gorodenkoff/Shutterstock)
Easy-to-use AI and automation platforms play a critical role here. When both technical and non-technical operators can design and deploy workflows, the pace of improvement accelerates. Not everyone needs to be an AI and automation expert, but everyone does need enough fluency to spot opportunities and act on them.
3. Create space for workflow redesign
A lot of organizations have built sandboxes where people can experiment with AI tools. That’s great, but it’s only half the invitation. The other half is creating space for people to actually propose and test entirely new workflows. This requires more trust, autonomy, and willingness to let someone challenge how a team operates.
Make workflow redesign a defined part of the work. Carve out time to revisit core processes, and invite the people closest to the friction—the ones doing the work day to day—to propose better ways of operating. Then invite them to run small pilots to test those ideas in practice.
4. Codify and share the best workflows
Workflow improvements often stay local. Someone finds a more effective way to work and keeps moving. The impact rarely extends beyond that one person.
Leaders can change that by treating workflows as shared assets. Capture what works in a simple format, whether that’s a quick walkthrough or a lightweight template. Once those workflows are visible, they start to spread. Others adapt and refine them in new contexts. What started as an individual shortcut becomes part of how the team operates. Over time, that shared layer compounds and raises the baseline for everyone.
The Advantage That Compounds
It was only a few years ago that most AI tools were limited to technical teams or locked behind waitlists. Today, they’re broadly available and increasingly easy to use. So the tools alone aren’t the long-term advantage. The human judgment to know where those capabilities belong, how to redesign work around them, and when to leave things alone—that’s the advantage that compounds.
The question that follows is straightforward, but it changes how you operate: who on your team can look at a workflow and see what it should become? Who can spot the steps that no longer need to exist, the decisions that can be made earlier, the work that can run without constant oversight?
The people who can answer that question are already on your team. Invest in them.
About the author: Emily Mabie is a senior AI automation engineer on Zapier‘s HR team, where she builds and enables AI-powered workflows that
help people teams work smarter. She draws on more than ten years in education and enablement to design tools and experiences that improve how people work, connect, and grow.
Related
The Limits of LLMs: Why Context, Not Models, Determines Success
Earlier this year, a wave of new AI coding capabilities crystallized a shift that had…
Frontier AI Model Power But 60X Cheaper? It’s Possible, Says Together AI
Frontier models like ChatGPT, Gemini, and Claude are very powerful, but that capability comes at…
Microsoft Launches New $2.5B AI Initiative With 6,000 Experts to Help Enterprises Deploy AI
Microsoft has launched Microsoft Frontier Company, a new consulting organization built to help enterprises plan…
The Agent Reliability Crisis
For years, the dominant concern for AI agents was capability. Could agents reason well enough?…
Databricks and Nvidia Launch Genesis Workbench for AI Scientists
AI is becoming an increasingly important tool for drug discovery. However, for many pharmaceutical and…
Hyperion Research Slices and Dices the HPC-AI Market
The market for HPC and AI for science solutions grew a healthy 17% in 2025,…
Broadcom Announces VMware Cloud Foundation 9.1
PALO ALTO, Calif., May 5, 2026 — Broadcom Inc., a global technology leader that designs,…
NVIDIA Announces Financial Results for 1st Quarter Fiscal 2027
SANTA CLARA, Calif., May 21, 2026 — NVIDIA (NASDAQ: NVDA) has reported record revenue for…
Cerebras Systems Announces Launch of Initial Public Offering
SUNNYVALE, Calif., May 4, 2026 — Cerebras Systems, Inc. today announced that it plans to commence…
What Stanford’s HAI Report Says About AI in Science
Progress in artificial intelligence continues to accelerate across a range of expert disciplines, according to…
Google Unveils Gemini Enterprise Agent Platform, Expands Vertex AI into Full Agent Stack
At Google Cloud Next 2026 in Las Vegas this week, the company announced major changes…
NVIDIA Launches Cosmos 3, the Open Frontier Foundation Model for Physical AI
TAIPEI, Taiwan, June 1, 2026 — NVIDIA today launched NVIDIA Cosmos 3, an open world…
IBM and Red Hat Expand Lightwell with New Offerings to Build the Trust Infrastructure for AI-Era Open Source
RALEIGH, N.C. and ARMONK, N.Y., July 8, 2026 — IBM and Red Hat today announced…
The Limits of LLMs: Why Context, Not Models, Determines Success
Earlier this year, a wave of new AI coding capabilities crystallized a shift that had…
Frontier AI Model Power But 60X Cheaper? It’s Possible, Says Together AI
Frontier models like ChatGPT, Gemini, and Claude are very powerful, but that capability comes at…
The AI Conversation Is Stuck On the Wrong Question. Here’s the One That Actually Matters
There are two ways to use AI inside a team. The first is the one…
LangChain and NVIDIA Launch NemoClaw Deep Agents Blueprint for Enterprise Agents
The NemoClaw for LangChain Deep Agents blueprint gives enterprises a reference architecture for building open…
Shanghai’s WAIC 2026 to Spotlight Global AI Governance and Industry Collaboration
BEIJING, July 8, 2026 — A news report from China Daily: The 2026 World Artificial…
