AI was promised to transform businesses from both the ground up and top down, creating new efficiencies, strategies, and ways of working. But the promise hasn’t yet been realized, and a much-publicized study from MIT found that 95% of companies are not realizing any returns from their AI investment.
Additionally, McKinsey found that 62% of companies are still experimenting and piloting AI, and haven’t yet found the right pathways to scaling. Business leaders are challenged with identifying the right use cases for AI, knowing how to apply AI to workflows and processes for maximum benefit, and understanding where to make the best investments in AI initiatives.
This is where a framework like Stanford University’s new Human Agency Scale can help business leaders better identify where AI is most and least applicable so they can make better investments and accelerate adoption. If you’re embarking on AI adoption but are overwhelmed by “all or nothing” implementation, or are unsure where AI fits into your organization, here’s how to use this framework to guide your decision-making.
What is the Human Agency Scale?
What if there were a more nuanced way to evaluate how much or little to use AI in an organization’s workflows and tasks? This question led researchers at Stanford University to create the Human Agency Scale (HAS), which “quantif[ies] the degree of human involvement required for occupational task completion and quality.”
HAS places tasks on a scale of “fully automated” to “fully human,” with varying degrees of automation or human augmentation along the way:
- H1: Tasks are fully automated, and the AI agent handles the task on its own.
- H2: Tasks are performed almost exclusively by the AI agent, with minimal human oversight, training, or direction.
- H3: Tasks are performed in partnership by AI and human.
- H4: Tasks are mostly completed by the human, with AI assistance; or, the AI agent can complete some of the tasks, but needs significant human guidance.
- H5: Tasks are performed by humans with minimal or no AI assistance; or, the AI agent cannot perform without continuous human oversight.
This scale now provides a common evaluative tool for organizations, as well as a common language when talking about AI and human performance or collaboration. It can also guide business leaders to more effectively implement AI and digital transformation initiatives.
How an AI Scale Leads to Smarter Implementation
Instead of taking an all-or-nothing approach to AI adoption, HAS or a similar framework can be used to evaluate the level of AI involvement in tasks and processes, which can inform larger AI strategies in the following ways:
Identifying use cases: With a framework like HAS, business leaders can identify the use cases where AI will function best in a more nuanced way than an all-or-nothing approach. For example, tasks that couldn’t be fully taken over by automation may have been dismissed in the past, but may be more suited for H3, H4, or H5 augmentation instead.
Guiding pilot programs: Identifying more nuanced applications of AI and automation can help guide smaller pilot programs. Instead of rolling out AI across an entire enterprise — hitting the hurdles and roadblocks that come with it — business leaders can better identify areas for experimentation, testing validity and impact before large-scale roll-out.
Sharper investments: Rolling out enterprise-level AI initiatives can be an expensive, daunting task for business leaders, especially in larger companies. By applying a scale to tasks to better understand the level of AI or automation needed, business leaders can make more exact investments in the tools and training they really need.
Clearer commitment: We’ve heard the stories where companies fully automated tasks and laid off staff, only to now back-pedal because it wasn’t the right decision. By evaluating tasks on a scale like HAS, business leaders can get clearer on their commitment to AI before making sweeping decisions that could impact revenue and reputation.
Improved employee agency and engagement: While more workers are beginning to understand that AI is a tool that will augment their duties, there are still many who fear AI will replace them. By identifying the level of AI use in each task, business leaders can more clearly communicate to their employees what will change in terms of AI integration and what will stay the same. This builds their sense of agency with AI use that can reduce fear of any AI threat, as well as increase confidence and engagement. If they need new AI training, upskilling and reskilling initiatives can build engagement and skills for the future as well.
Better aligned AI strategy: By taking the guesswork and assumptions out of where and how to implement AI, business leaders can create more aligned AI strategies, from the boardroom and back office, to the front lines.
A More Aligned AI Strategy — and Proteus Can Help
It’s one thing to present a better way to evaluate tasks for AI and automation use, but it’s another to know how to go about evaluating those tasks and translating those findings into actionable process improvements.
That’s where Proteus can help. In our experience working with organizations on their AI transformation, we see four common roadblocks that stop organizations in their path forward:
- Leadership lacks a coherent vision for how they want to use AI in their organization, and without it, AI often lands with the IT team as another “innovation.”
- Leadership knows there’s potential for AI, but isn’t sure what to automate vs. augment or where to put their focus, which leads to constant pivoting and more manual work than needed.
- The organization is using vendor-built AI agents or digital employees, but is having trouble formally integrating them into their processes or institutional knowledge.
- The organization is already experimenting with AI, but they’re staying in the demo stage, with little to no structured adoption.
If you see yourself in one of the roadblocks above, we can work with you turn your AI ambitions into reality by:
- using HAS to classify your critical tasks, so that you can see exactly where AI benefits your processes and where it hinders them.
- defining your AI vision and how to get there by identifying process friction points and outdated mindsets holding your organization back from AI transformation.
- remapping your processes, from decision-making to supply chain, so they reflect your new AI optimization plan.
- creating new internal change stories, as well as identifying AI ambassadors throughout your organization to move your new AI initiatives forward.
- increasing confidence that you’re making the best choices, investments, and forward direction in your AI initiatives.
AI’s promises to transform businesses from both the ground up and top down aren’t empty. Business leaders just need better ways to evaluate how they apply AI in their organization and where investments are going to pay off the most.
If you’re ready to learn how to implement a framework like HAS to inform your next AI steps, let’s connect.
Want to learn more about HAS and how humans can better collaborate with AI? Check out our podcast episode “From Experiments to Impact: Building Human-AI Collaboration.”
Never miss a new insight from us! Subscribe to our newsletter today for the latest on how to lead true transformation in your organization.
