Featured on FORBES: Human-AI Collaboration: Transitioning From Automation To Augmentation
We are moving from a period of AI-powered automation to augmentation, from replacement to enhancement and from mechanization to collaboration. As we bump up against the limitations on the current generation of AI tools, employers are reconsidering the roles of humans and machines, and how these roles can complement one another. This shift could bring optimism to human workers and more clarity on the best ways to leverage AI tools for businesses.
The Fear Factor
Anxiety over AI is real. Workers are afraid of being replaced by machines and that fear is not totally unfounded. In 2025, consulting firm Challenger, Gray & Christmas reported that employers cited AI-powered automation as the reason for cutting an estimated 55,000 jobs. The vast majority of these jobs were concentrated in areas like media, IT and telecom. But there was another employment trend. During the same period, the firm found that employment in healthcare, education and hospitality grew. The need for human talent has not diminished, but is shifting. And as our needs shift, the way we think about and leverage AI-powered technology must shift, too.
In the years since the release of ChatGPT, we have seen both the promise and the limitations of generative AI. The current generation of AI tools has demonstrated capacity for speed, efficiency and force multiplication. It has also shown vulnerability to bias, inaccuracy and manipulation. Yet researchers are increasingly optimistic about the role of human ingenuity in the AI era.
The EPOCH Framework
For those of us nursing latent dread about the eventuality of replacement, a recent study from MIT Sloan School of Management may offer some cause for hope. The study found that as AI tools evolve, the technology will serve an increasingly complementary role in human work.
In their study, MIT professors provided statistical analysis of tasks that AI is bad at. Subsequently, they identified the different strategies and actions that human operators used to address and compensate for these limitations. From there, the study identified a set of distinctly human skills that will be of increasing value in an era of human-AI collaboration.
They called this framework of skills EPOCH:
- Empathy and Emotional Intelligence
- Presence, Networking and Connectedness
- Opinion, Judgment, and Ethics
- Creativity and Imagination
- Hope, Vision and Leadership.
Essentially, tasks that do not require these qualities can be automated and jobs that do not demand these qualities are at the highest risk of replacement. But in areas where these traits are required, it’s worth asking the question: How can AI tools augment distinctly human work?
Recasting Rather Than Replacing
It may come as no surprise that executives and workers see things a little differently. BCG’s 2025 AI Radar report indicated that 75% of key decision makers see AI as a strategic priority—a sharp contrast to the 52% of workers who worry more about the future impact of AI on human work. The good news is the majority of executive see humans and AI working side by side and leaders are focused on upskilling their current workers to ensure that we’re making the most of what these potentially powerful tools can do.
But this upskilling must also be done with a clear understanding of what these tools can’t do. I’ve seen (and written about) plenty of stories involving AI implementation failure. A March 2026 article from Business Insider, “The Great AI Deskilling Has Begun,” suggests that some of these failures have been due to an overinflated sense of what AI technology promises. This can contribute to worker dependency and essentially deskill workers. In short, the article argues that we’re surrendering ingenuity and creative thinking to these tools. And not only are we getting disappointing results, but we’re also finding it’s difficult to recover these skills.
Retaining Human Skills
So, how can you ensure your workers are using the skills that differentiate them from robots?
Upskill your workers.
Provide training for AI fluency, including tools needed to review, judge, interpret and modify AI output.
Teach human-in-the-loop orientation.
Use AI for predictable, repetitive and data-heavy tasks while humans guide machine output.
Prioritize soft skills.
Elevate teamwork, storytelling, creativity, empathy and leadership in recruitment and role assignment.
There are already real-world examples where this collaboration is enhancing human work. In healthcare, physicians are using the assistance of AI-powered tools to improve diagnostic accuracy while still doing the work of patient care and treatment planning. In the legal profession, firms are using AI-powered tools to comb through enormous volumes of case history in seconds while attorneys focus on arguing the case at hand. Even in software development, where programmers have been hard hit by AI replacement, developers are finding more time to focus on architecture and system-level design while AI tools do the coding, testing and debugging.
Asking New Questions About AI
We must transition the way we think about AI. In the earliest stages, it has been our natural impulse to ask, “Can we automate this task?” Today, it may make a lot more sense to ask, “Should we automate this task?”
For businesses asking the latter question, they are evolving their mindset toward AI. More employers are coming to view AI as a way to strengthen human work rather than replace it. This should come as a relief to human workers, but it also challenges all of us to accept and better understand these tools or risk our own obsolescence.