Here»s a thing – although most C-Suite leaders are drinking the Kool-Aid that generative AI will boost productivity, their employees are not so sure. In fact, just over three quarters (77%) believe the technology simply adds to their workload, according to a study by freelance platform, Upwork.
Some of this is because they have to spend time learning how to use the new systems (23%). But two in five also complain of needing to spend more time reviewing or moderating AI content. One in five even say they are finding themselves being asked to do more as a direct result of the technology being introduced.
To make matters worse, 47% of employees have no idea how they are supposed to achieve the productivity gains desired. Two out of five believe their employers are asking too much. As a result, one in three plan to quit in the next six months. Want to hear more about Generative AI and LLMs?
We’ll send you our top articles (and no marketing spam), no more than once a week. Email * Name *
What is going wrong?
So, why do things appear to be going so wrong and why are so many organizations apparently experiencing the exact opposite to what they had hoped for in productivity terms?
There appear to be a number of issues at play here, some old, some rather newer. Dr Steve Clapp is a member of the Human Factors and Ergonomics Society and Post-doctorate Scholar at The Readability Consortium at the University of Central Florida. He explains:
What’s been happening for years is that some shiny, new thing in tech catches the eye of leaders, and they think it’ll be cool to implement. So, they throw it at employees and tell them what they want to happen. But people are given little instruction in how to use it and, more importantly, how it aligns with, or gives meaning to, their work. So, on the face of it, AI may seem to make sense. But there frequently isn’t enough explanation about how to use it, what its benefits are for employees and how it’ll make their lives easier. Also, people often fear it’ll be used to replace them, so they don’t buy into it.
But this emotional buy-in is vital for any kind of technological initiative to succeed, Clapp believes:
The employee piece of the model is usually either given lip service or not fully implemented. Instead, systems are built on showing return on investment in a relatively short time.
Humans are struggling to keep up
Teresa Ramos is an Executive Coach, Digital Consultant and Founder of globalresults. She agrees that employers are repeating the same implementation mistakes with Generative AI that they have made for years with other technologies. But she also believes the situation is currently being made worse due to the speed and level of change taking place. Moreover, Ramos adds, there tends to be too much focus on productivity and efficiency gains rather than putting people at the heart of such change. She says:
Things are shifting so fast that we as humans are struggling to keep up with it. It’s about the velocity and scope of change, but AI isn’t only the latest issue here. Previously it was Big Data, followed by digital transformation. So, employees feel tired, overworked and overwhelmed. They don’t know how AI works and just see it as something new to learn. They also don’t understand the benefits it’ll bring them.
Another problem is that people often feel they are expected to use the technology proficiently from day one and are concerned about what will happen if they make a mistake. As a result, Ramos says:
People go into fight or flight mode, so to implement the technology effectively, these cultural issues have to be managed. With digital transformation, it was 20% about digital and 80% about transformation, and it’s the same with AI.
Amalia Goodwin, Managing Director of professional services and business consultancy Slalom’s Global Advisory Practice, takes it a step further. She believes that to implement Generative AI effectively, it is necessary to change the way people work:
To find true value from an organizational standpoint, it’s not just about making things go faster. It requires taking a people-centric rather than tech-centric approach based on strategy and what you want to get out of work. So, you may need to rethink it and how it’s performed. You can’t just expect employees to figure things out for themselves. It’s important to look holistically at workflows and how you want to get things done. This means it’s about being more structured in your approach and working on these things as a team. This isn’t the remit of one person – everyone has to be involved.
The changing nature of work
The first step in going down this route is for organizations to understand what their business goals are and what they want to achieve. Then it is about analyzing processes and workflows to identify bottlenecks and areas for improvement. The next stage is to work out what role automation and AI on the one hand, and humans on the other, should play. Goodwin explains:
So, you need to take a step back and look at your overall strategy and the interventions required to see where AI and automation could make a difference. For example, by creating more strategic roles that require human critical thinking, you might find it’s possible to hire two employees to replace the three who are retiring and automate the rest. It’s about thinking how work gets done and the skills and tasks required to do it, so you know where to focus precious resources. If you’re doing it right, you should make work more human. But it takes time and strategy, and you certainly can’t just turn on the technology and think wonderful things will happen.
Deepika Adusumilli is Chief Data & AI Officer at telco, BT Group. She believes there are three key stages to getting AI initiatives right. The first is identifying appropriate use cases and devising relevant key performance indicators and metrics to evaluate success. The second involves employee education and training. Possible approaches here include gamification and group learning. Just as vital is the introduction of change management practices to win people’s hearts and minds and ensure buy-in. As Ramos explains:
You have to involve employees as they’re the ones who can tell you if the optimization you have in mind is achievable, and it also helps overcome resistance. But people require training too as well as the space to adapt, learn and make mistakes. So, it’s important to factor in not just training costs, but also adoption time. Only after that will you see productivity gains. It doesn’t happen at the push of a button.
Due to the immaturity of generative AI though, she adds, budgeting for a process of continuous learning is also crucial to ensure employees do not fall behind over time.
The third consideration, meanwhile, is to understand whether organizations’ chosen technology will provide the expected outcomes. This is particularly important if, as is the case with Generative AI, it is still evolving. As Adusumilli says:
Generative AI is very interesting, but it must be used in a way that brings value to your data. The technology works on unstructured data, which means it has a vast number of potential use cases. As a result, it can be so transformative that it needs more than the usual amount of change management on top. The basics are still the same, which includes putting proper governance and metrics in place. It just needs even more focus than usual.