Authors: Nathalie Francisci Michelle Moore
Myth or fact: Canadians are more risk averse than their global counterparts.
Many have debated the Canadian subject, including those noting risk aversion has served the country well at times (recall the 2008 global financial recession). However, with technology, risk aversion is generally an obstacle to innovation and global competitiveness.
OpenAI's introduction of generative artificial intelligence (GenAI) in 2022 in the form of ChatGPT became the biggest tech disruptor since the internet. Those bold enough to embrace it with appropriate guardrails gain a competitive edge. Yet, in a November 2023 IBM survey of companies in 20 countries, Canada was the country least likely (35%) to accelerate the rollout of GenAI.1
Risk averse or not, Canadian employers see GenAI as a hot topic. Sixty-three percent of organizations are rethinking the role of their HR departments because of the entrance of AI in workplaces, according to a 2024 survey by PwC, a publication of KM Business Information Canada.2 HR leaders directed to oversee the organizational-wide adoption of GenAI may feel constrained by limited technological expertise. Further, HR teams may have had little or no opportunity to weigh in on strategies for helping employees adapt and adopt this new technology before its rollout.
At Gallagher, our conversations with clients reveal various reasons for the slow adoption rate of GenAI. Examples include fear of risks and loss of control, poor initial user experience, and lack of leadership and guidance. This article examines these factors and offers a high-level look at the issues HR leaders must consider to help drive GenAI adoption safely, support organizational goals and face the future with confidence.
Where is your organization in adapting GenAI for HR use?
Currently most organizations fall in roughly equal numbers into one of three buckets of adoption:
- The organization prohibits using GenAI in the workplace. This policy may stem from a board or investor group that wants to see value creation before supporting adoption. Nonetheless, individual employees, especially younger generations, commonly use personal devices to access GenAI tools such as ChatGPT when doing their work.
- The organization permits using open (public) GenAI for work. Organizations see mixed results, including an increase in efficiency and productivity coupled with problems associated with security and quality control, among others.
- The organization uses a closed (proprietary) system to mitigate risk. Although more costly, proprietary systems provide a safe and secure environment, reducing risks and misinformation associated with open systems. Closed systems are critical if the training dataset includes sensitive employee data.
Adoption often remains low in the workplace — even where the employer provides a propriety closed system. Especially, small to midsize organizations fail to provide the guidance and guardrails necessary for a productive and positive experience that encourages adoption. Conversely, mature — typically larger — organizations report higher GenAI adoption rates.
Overcoming adoption barriers
To accelerate user adoption of GenAI, organizations first must understand the most common barriers.
Fear. The fear of GenAI technology associated with risks and loss of control is understandable. The complexity and rapid evolution of the technology make it difficult to counter this fear, as do expectations from board and investor leadership who may look for immediate returns on the investment or can't see a path to mitigate risk.
Poor user experience. As with learning any new technology, it's messy, so practice and patience are required. A safe experimentation and shared learning environment will rapidly improve the user experience. Such an approach also will improve the technology. GenAI needs to learn your organization, just as employees need to learn how to leverage GenAI. Increased engagement advances both human and AI learning.
Lack of guidance on usage, including guardrails. When organizations throw open the gates with no guardrails or guidance, the result can be chaos, frustrating employees and leadership.
In part 2 of this series, we offer strategies and tips for employers to counter these three barriers through a change management strategy in partnership with IT. Before embarking on an adoption strategy, however, HR leaders need a basic understanding of the technology. The more you understand GenAI benefits and risks, the better you can advise the C-suite and board on safely introducing GenAI and drive user adoption to support organizational wellbeing.
What is — and isn't — artificial intelligence?
AI imitates human behavior. The technology relies on algorithms and an immense amount of data to respond to queries. Many confuse hyper-automation and machine learning with true AI, exemplified by tools like ChatGPT and Microsoft CoPilot. Whereas machine learning identifies patterns within a training data set to make predictions or decisions, GenAI creates new content based on existing data and machine learning algorithms.
GenAI has significantly sparked employers' interest in and use of true AI. The buzz centers on opportunities to increase productivity by freeing employees to focus on strategic initiatives by offloading transactional functions that the technology can safely and efficiently manage. Still, you may have heard horror stories about GenAI associated with misinformation ("hallucinations") and bias. Among our favorites is the lawyer who used ChatGPT to cite case law, only for the judge to point out that the cited cases didn't exist.
Understanding the potential impact of non-human content creation is essential for chief human resources officers (CHROs) advocating organizational GenAI adoption. "AI-generated misinformation/disinformation" ranked second on a list of top global risks in 2024 in a survey of organizational leaders conducted by the World Economic Forum. Respondents ranked it No. 1 when asked about the likely impact (severity) in the next two years.3
Big risks: Faulty facts and bias
Beware of unintended bias
In 2018, Amazon scrapped its AI recruitment tool after finding the tool discriminated against female candidates. The unintended bias lay not in the tech but in the training method of feeding AI resumes of past "successful" candidates, which were predominately male. If diversity, equity and inclusion (DEI) is an organizational priority, proactively define what "success" looks like in your organization and continue to monitor and adjust internal processes and practices.
Never assume AI-generated information is accurate. Many generative AI tools cite references, so review all sources. Often, the cited source isn't the original source. Follow the data trail to its origin. When that's not possible, evaluate the quality and credibility of the cited source. For example, was the source of information CBC News or a Society for Human Resource Management (SHRM) article versus an unknown blogger or product-sponsored survey?
GenAI outputs reflect the bias of the information used to train the technology — information created by inherently biased humans. GenAI bias, subtle or overt, can color your brand when used repeatedly in communications. Still, with human oversight, you can counter bias.
Manage risks with experimentation and guardrails
The best results from GenAI tools come from crafting effective prompts. A poorly worded prompt will return, at best, incomplete or off-point results. At worst, it may contain false information presented as organizational policy or guidance. Consider these best practices, informed by ChatGPT, for asking for information that will return the best results and reduce your risk.
Be clear, specific and grammatically correct in your prompt. State the topic and what you want the tool to do. Break down a complex subject into multiple narrowly focused prompts. Avoid shortened words or phrases as used in texts.
Provide context. GenAI will try to create context if you don't provide it, potentially leading to false or irrelevant information. Start with "based on the information below," then paste the information.
Include hypothetical examples. GenAI learns from your prompts, so provide real-world scenarios to support your queries.
Don't ask leading questions. If your prompt assumes a particular viewpoint or answer, expect to see it reflected in the response. Be aware of prompts that could lead to biased results.
Prompt. Assess. Repeat. Practice crafting prompts for a topic you know well. Assess the results for what is wrong or missing and give it another go. Like GenAI, you will learn by doing.
Closed AI systems for organizational data
ChatGPT, CoPilot and other GenAI tools learn from massive amounts of unlimited and uncontrolled input for the training dataset. These open systems are available to many users who can add new data and change the model to suit their needs. Open systems foster collaboration and rapid innovation.
Conversely, in a proprietary closed system, the owner controls the training dataset, which emphasizes control, security and competitive advantage. A closed system is a must-have if the training dataset uses sensitive employee information.
While hybrid models are emerging, both open and closed systems offer pros and cons. Whether open or closed, GenAI requires employer guardrails to help employees understand the appropriate and approved use for work purposes. A 2023 Gallagher survey of US organizational communicators found that only 29% of organizations provide employees guidance on when, where or how to use AI, and just 20% provide AI training or resources.4 Organizations that have developed their own AI solution tend to apply better governance and oversight than those that use an open AI tool.
Experimentation is key
Experimenting with GenAI in a safe environment is key to learning to use the technology effectively and to minimize risk. A December 2023 Gallagher poll of US employers found that the top applications of AI within HR include creating emails, recruiting and writing job descriptions.5 All are low-risk functions to begin experimenting and practicing crafting prompts, assuming a human reviews the outputs for accuracy, bias and preferred tone before using in the real world.
Practice crafting prompts can help move the needle on adoption. The Gallagher communications survey found that those using or experimenting with AI see its benefits and feel more positive overall about the value of AI — 55% compared to 45%.4
AI is coming to your organization
As of early 2024, roughly 14% of Canadian businesses are engaged with or plan to use generative AI technologies. Adoption rates are higher in some sectors, notably IT and cultural industries, where 24% of businesses report using it. Organizations in the professional, scientific and technical services sectors rank second for use (18.8%) or plans to use (9.5%), followed by finance and insurance, with 16.9% currently using it and 5.8% planning to use it.6
Despite what may seem like low numbers of companies implementing GenAI, it's simply a matter of time before more employers adopt the technology. A 2024 study by Deloitte found that 64% of CEOs feel pressured to accelerate GenAI adoption due to investor and competitive demands.7
A Gartner survey of HR leaders found that 38% are actively piloting or have implemented GenAI technologies, up from 19% in a survey conducted just six months earlier.8 The clock is ticking, and HR leaders must prepare.
HR's value within the organization will increase substantially if you understand the technology, its benefits and risks and ways to use it effectively and safely. However, introducing GenAI into HR and other organizational processes doesn't guarantee user adoption.
Further, the onus for user adoption cannot sit with HR alone. The CHRO's charge is to maximize organizational performance using human capital. The CIO's charge is to maximize organizational performance using technology. HR and IT must collaborate, combining their expertise and insights to advance GenAI user adoption.
Part 2 of this series examines an HR-IT collaborative strategy for technology user adoption and offers a practical, hands-on approach to GenAI technologies.
Gallagher can help
Our team of HR consultants is here to help you explore the potential benefits of integrating AI into your HR strategy and practices. We offer personalized advice and tailored solutions to meet your organization's specific needs. Start the discussion and gain powerful insights.
