Generative AI (GenAI) is now commonplace in many workplaces, with tools like ChatGPT, Claude.ai, Jasper, and Co-Pilot to assist with tedious tasks and increase productivity across departments. However, even in cases of widespread adoption, it remains challenging to achieve a tangible return on investment (ROI) from GenAI. The problem might not lie in the AI tools themselves, but rather in how and why they are being adopted within the workforce.
According to a McKinsey report, only 20% of companies are seeing an impact from AI on their bottom line.
Even though GenAI offers significant business benefits, leadership is not effectively conveying these benefits downstream, which means their organizations struggle to move beyond pilots and encounter weak strategies for widespread adoption and sporadic skills training gaps. A clear framework is needed to unlock the true ROI behind the metaphorical AI curtain.
Advocating for GenAI Adoption: Three Tactics
Here are three strategies to encourage widespread GenAI adoption within your business:
- Tackle the Shadow IT Issue Upfront
McKinsey found that 71% of employees use GenAI for at least one business function. Why is it so hard to quantify ROI on AI usage if this is the case? There could be a few reasons at play: for one, the employees might not be using GenAI tools through an official channel and therefore the ROI is not tracked. Another reason could be that the GenAI prompting is not on par with expectations, which could lead to haphazard usage and not continually investing in pattern development to reap more meaningful outputs.
The adoption patterns we’re seeing with GenAI are like what we’ve seen before with early cloud adoption. Many highly skilled workers remain skeptical of GenAI’s value and therefore, the true adoption is occurring in a bottom-up approach rather than embracing the AI tools the organization’s leaders have passed down the chain.
Much like shadow IT organizations encountered during the emerging cloud computing era, employees were using the cloud, but not the cloud tools the organization wanted them to use. This approach can lead to cybersecurity and compliance risks, so IT teams should be aware that unrestricted GenAI usage is likely already occurring at their organization. Keep in mind, too, that it may even present a legal and privacy risk, given that machine learning models retain the information fed to them for further pattern recognition — what are employees sharing with the AI companies?
When employees seek to use tools like GenAI through unofficial channels, it exposes the insufficiency of current IT tools. On the other hand, some IT departments may be implementing tools like SurePath.AI that block the usage and modify prompts to exclude sensitive information—if this is the case, it likely hampers employee workflows.
It’s not a time to scold but rather to have an earnest conversation with your workforce. What do they need to truly be successful? Re-evaluate which solutions are available to ease the transition into modernized technology usage. Flipping the script to embrace these new technologies — just as in the age of cloud computing — at least brings the shadow IT problem into the light, which allows for proper (and fast) mitigation.
- Create a Foundation for the Technology to Grow
The age of GenAI is similar to the age of cloud adoption, when several departments were selecting the cloud models that suited their department’s needs rather than establishing a cohesive tool for their entire organization. It resulted in a wild west scenario. When it comes to GenAI, a business’s departments may end up using several solutions with overlapping capabilities, which wastes organizational budget and lacks a clear, cohesive strategy for the long term.
Correcting your business infrastructure and IT foundation can correct the shadow IT challenge with GenAI. To do this, you will need to align the organizational tech stack to fit business strategy objectives. This demands a conversation at the highest level to confirm the most essential must-have elements, making decisions on what and where to cut much easier.
Define the metrics that GenAI must meet to be considered a success at your business. When employees approach your IT team for new apps, this high-water mark will be a more straightforward yes/no answer. If the current apps aren’t meeting the mark, they must be evaluated for continued usage.
- Educate Your Workforce (and Don’t Just Hire New Folks)
Training is quickly becoming the most significant piece of the puzzle that many businesses neglect. More than half of the respondents in this article from CIO Dive reported having the right talent for their AI initiatives. This means competition for already-trained AI folks is fierce.
The answer is not just hiring new talent to fill the skills gaps in your business. If you’re not actively educating your existing workforce on how and when to use GenAI tools, you’ll soon find that only your new hires can keep up. For your existing workforce, it’s key to make the case to them for why AI change is needed, as there are always folks resistant to change, to ensure you don’t experience a wide exodus. If attrition rates get out of hand, you could be dealing with an IP loss problem while simultaneously trying to accelerate productivity and business revenue.
Think of how IT teams moved from traditional waterfall project management to agile methodology: it takes time to get everyone on board, especially in an era of AI transformation. People have to see that change is a good thing and necessary for their careers.
AI Readiness Depends Upon Strategy Readiness
Aligning with your core stakeholders (executive and board leadership, middle managers, IT team and HR employees) about how and when to use AI — and how and when to educate your workforce on these aspects — is essential to tackling the age of GenAI with confidence and a balanced budget. ROI will be impossible without these elements.
Setting the foundation for success means having the right IT foundation for acceleration and fast-paced pivoting. Without having the right conditions already in place, we’ll continue to see ill-fated adoptions and risky GenAI uses.
