Two Simple Questions That Can Improve Your AI Adoption Rate

Read transcript

The widespread rush to adopt artificial intelligence in workplaces has been nothing short of transformational. But in the course of pursuing that transformation, a crucial question often gets overlooked:

Are you automating the right tasks — or just the ones that seem the easiest to use AI for?

If you’re not sure, that’s OK. It’s not too late to pause, figure this out, and course correct. 

The two questions that bring AI disconnection to light

Many organizations are investing heavily in AI solutions with the promise of increased productivity, lower costs, and faster workflows. But there’s a growing disconnect between what companies are building and what employees actually want. 

It’s a situation that’s familiar to business leaders everywhere: they’re pushing to adopt AI, companies are getting on board, and … usage remains spotty. Some workers are bypassing the tools all together in favor of standard manual processes, while others comply reluctantly. Some may even verbally push back, saying their work is now harder or slower. They may say that AI shouldn’t be doing certain parts of their jobs. 

And honestly? They’re probably right. The disconnect that’s happening right now isn’t due to a technical problem; it’s an organizational issue.

Stanford University researchers analyzed 844 tasks across 104 occupations and assessed each one using two lines of questioning: 

1. Do workers want to automate this task?
2. Can AI perform the task well from a technical standpoint?

Through this analysis, the research team developed four distinct “task zones.”

4 task zones that lead to better AI adoption

The four task zones are simple and straightforward. When applied as a framework to guide your business operations, these zones make it much easier to figure out exactly what should and shouldn’t be automated.

1. Automation green light zone: high desire, high capability

“Green light” tasks are ones where automation aligns with both employee enthusiasm and technical capability. Repetitive, manual work typically falls in this zone — things like scheduling, document tagging, and data entry. 

Many workers are happy to hand off this type of work to AI, and the technology typically produces a usable output. 

The green light zone offers the highest potential for productivity gains and smoother adoption without burnout. 

2. Automation red light zone: low desire, high capability 

The “red light” zone is where many organizations go wrong. Tasks in this zone are well-suited to automation … but workers don’t want to outsource this work to a machine. 

Red light zone tasks are often creative and very human in nature, including:

• Strategic brainstorming
• Creative ideation
• Mentoring 
• Interpersonal communication 

3. R&D opportunity zone: high desire, low capability 

This zone is full of tasks that employees want to automate, but the technology isn’t quite ready to do so. Think real-time project reprioritization and nuanced decision-making that adapts to shifting context — work that requires a human touch, at least for now. 

Recent advancements in agentic AI may mean that some tasks in the R&D zone become “green light” opportunities in the near future — but continued innovation is needed.

4. Low priority zone: low desire, low capability

Low priority zone tasks are ones that people don’t want to automate and AI can’t execute well. These can include context-rich, highly personalized things like:

• Nuanced performance reviews
• Complex negotiations 
• Culture-building conversations and experiences

Trying to force automation in the low priority zone leads to wasted effort. 

The true cost of AI misalignment

The Stanford paper shows us that the problem isn’t that AI tools exist — it’s how and where they’re being used. Even the most advanced automation can backfire when it disrupts the wrong parts of someone’s job.

When organizations prioritize automation based solely on technical feasibility, without considering employee sentiment, they risk more than losing money on poor adoption. They invite active resistance, diminished morale, and higher rates of burnout. 

Research from our From Tools to Teammates report underscores this tension — we found that the workers who reported the highest productivity gains from AI were also the most likely to report feeling burned out. A staggering 88% of these high-AI performers also reported burnout symptoms.

This misalignment is especially damaging when AI targets the “red light” zone, the human-centered work people don’t want to give up. These tasks aren’t only fulfilling; they’re also what differentiate strong teams in complex, fast-moving environments.

Human skills are power skills

As AI increasingly handles information-heavy, rules-based tasks that fall into the “green zone,” it’s redefining which skills hold the most value at work. Research from both Stanford and Upwork points to the same conclusion: Roles that rely on human nuance, not repetition, are rising in importance.

The job market is shifting toward roles that emphasize uniquely human traits like: 

• Emotional intelligence
• Conflict resolution
• Team leadership
• Systems thinking

These aren’t just "nice to have" traits, either; they’re essential power skills that support high-performing and adaptive teams. And they aren’t skills that can be easily automated.

This shift should reshape how organizations think about training. Upskilling can't focus solely on technical know-how like prompt engineering or tool navigation.

Nurturing capabilities like empathy and collaboration helps organizations get more from AI while keeping people at the center. It's not just about adopting smarter tools. It's about enabling teams to use them wisely — combining technology with human insight, creativity, and collaboration.

3 ways to better align AI automation with workplace needs

Technology should serve people — not the other way around. To make AI work in a way that benefits both your teams and your bottom line, organizations need to get intentional about how tools are chosen, tested, and communicated.

Here are three practical strategies to close the gap between AI potential and real-world adoption.

1. Listen before building

Before launching any new AI feature or tool, talk with the people it’s meant to help. Use surveys, interviews, or retrospectives to ask questions like:

• What tasks drain your energy?
• Where could AI provide meaningful support?
• What parts of your work do you value most and want to keep?

The goal is to uncover pain points — and enjoyable tasks — from the user’s perspective, not just technical friction or IT requirements.

2. Start in the green light zone

Instead of guessing where AI will have the most impact, begin with the tasks that your employees want to automate. For example, you might test an AI assistant that summarizes meeting notes, not one that plans creative strategy for your next marketing campaign.

When employees immediately experience how a tool saves them time or reduces friction on tasks that they want to automate, adoption tends to grow organically — no extensive change management required.

3. Reinforce the value of human skills

Don’t let AI adoption send the wrong signal. If employees feel like interpersonal or judgment-driven work is being devalued, morale can suffer. Make it clear that skills like decision-making, creative strategizing, mentoring, and culture-building are still core to your team’s success.

Frame AI as a support tool that frees people to do more of what they do best — not a replacement for their expertise.

Remember: Start small, then scale to bigger solutions

Rolling out AI tools without employee input is a risky move — and increasingly, an outdated one. 

Leaders don’t need massive overhauls to get this right. Even a short, targeted survey can reveal critical insights about tool adoption and workflow friction. Start with one team and identify their green light zone — then build from there, aligning AI development with what your workers actually need and want. 

Similar principles apply for individual contributors and freelancers who want to add more AI automation into their work. Ask yourself whether you're using AI to offload the right kinds of tasks — or if it's creeping into the parts of your job you actually enjoy. Doing so can help you reclaim space for strategy and creativity, while still getting boring, repetitive tasks done in the background. 

Create your human-centered AI strategy

The bottom line: AI adoption doesn’t just depend on what’s possible. It depends on what’s welcome. And that starts by asking the right questions.

Organizational development consultants, change management specialists, and AI engineers can help you plan and build the right AI strategy for your organization. You can get help from all of these professionals right now through Upwork — sign up or log in to get started.