The IT/ERP community is experiencing a hype around generative artificial intelligence: Workday's AI approach is different.

According to a recent global study, a significant proportion of management and employees already see the potential benefits of AI: 62% of managers (C-suite or those reporting directly to them) and 52% of employees rate the development as positive. However, there are also numerous negative voices stating that AI could replace certain jobs and therefore employees, that there is a loss of control over AI in the company or that there is simply too little information about the risks. To counteract these fears and promote a responsible approach to the use of AI, we at Workday are working with our customers, partners and legislators to jointly develop and implement binding and ethical guidelines for the use of AI.

The report entitled "Preparing to Power Up: EMEA is leading the way to an AI-driven future" shows that company managers are fundamentally convinced of the benefits of AI and ML, even if opinions differ in detail in the boardrooms. However, it is also clear that data silos and bureaucracy stand in the way of a comprehensive introduction. The Workday survey also shows that EMEA business leaders are generally more confident in artificial intelligence and machine learning than executives in other regions. Following the introduction of the General Data Protection Regulation (GDPR), the European Union is currently planning an AI Act to set out the legal framework for the use of artificial intelligence. The establishment of AI governance is intended to ensure data quality, transparency and human control in order to achieve broader acceptance of the new technologies.

"The EU AI law is a milestone and a remarkable step forward in the responsible use and development of AI," said Jens-Henrik Jeppesen, Senior Director of Public Policy at Workday. "At Workday, we've been building AI and ML capabilities for nearly a decade. We know that these technologies need to gain people's trust to realize their full potential. Smart, risk-based and proportionate regulation can help build that trust and avert the risk of potential harm, while enabling innovation. We hope that the new EU AI Act will do just that when it comes into force and encourage responsible use of these technologies by companies." The combination of clear regulations and the widespread enthusiasm for AI and ML among EMEA leaders is leading to a noticeably higher willingness to invest. Northern European players are in the lead here, while companies in Southern Europe prefer a more conservative investment approach.

Ethical approach as a basis

AI can not only add real value, it can also make a difference for companies in terms of productivity and efficiency in international competition. However, a key prerequisite is that AI is perceived as trustworthy and understood as a meaningful addition to human experience and perception. This can only succeed if there is a clear understanding of the development and evaluation of the AI and ML products supplied and that their use is based on compliance with ethical AI principles. This requires close collaboration, clear guidelines and documented procedures. Practical, risk-based regulations increase confidence in AI technology and facilitate innovation.

When using generative AI, Workday focuses on complementing rather than displacing human potential. Rather, our approach includes collaboration between humans and machines to further integrate human expertise. Through our AI practices, we are helping to ensure that human judgment remains in its central role as the ultimate decision-making factor. Back in 2019, we committed to an ethical approach to AI, with employees, innovation and integrity at its core. We communicate transparently about our approach, including the development process, application principles and our commitment to ethical AI standards. 

In addition, Workday is taking a leading role in shaping AI-focused discussions at various levels of government: In the United States, we actively participate in policy dialogues at the federal, state, and local levels to advocate for thoughtful and effective regulations for AI applications. Beyond national borders, we have built strong partnerships with the European Union and other global governments, working together to develop comprehensive policy approaches that promote the responsible use of AI worldwide.

Difference in the IT architecture

But how does Workday differ from other providers? A key difference lies in the IT architecture: all customers work with the same version of Workday - including the same data model. At Workday, we have over 65 million users and around 629 billion transactions processed by the system every year - and the number is growing. With the permission of our customers, we use this high-quality data for AI functions. It enables us to create models that
deliver accurate, meaningful and trustworthy results based on the collective insights of our customer community.

And unlike other providers, this gives us an important differentiator, because our AI is based on a unique data set: the effectiveness of AI depends on the quantity and quality of the data on which it is built. At Workday, we have also been relying on specialized, large language models (LLMs) to drive generative AI for years.

We are currently building capabilities that leverage generative AI for various language and image-related tasks, including natural language generation, document understanding, and content search, summarization and augmentation. We also follow our platform strategy when developing and deploying generative AI: we can quickly and easily leverage new technologies to develop new capabilities while ensuring a consistent experience across the Workday environment. And this is another difference, as Workday is able to quickly implement new models or reuse existing models for extended use cases.

In the case of generative AI, we take a hybrid, vendor-agnostic approach - developing our own domain-specific LLMs on the one hand, and working closely with leading third-party vendors to create mixed or shared models on the other. This approach allows us to leverage the best available technologies while providing our clients with powerful, cost-effective and trustworthy solutions.

In addition to the different pace of implementation, AI and ML technologies are also slowed down by other hurdles. A lack of data management and excessive bureaucracy are the main reasons why company managers are still hesitant to introduce AI and ML. 60 percent of EMEA companies complain about data silos that make it difficult for them to gain relevant insights in real time. In many companies, a significant cultural shift needs to take place in order to adopt AI and ML on a larger scale. Without comprehensive support from top management, this is doomed to fail. Here, too, the Workday survey reveals a problem: 37 percent of EMEA leaders believe that their management has not yet realized the importance of AI and ML.

Managers and employees alike want people to have a significant influence on AI-controlled processes. What exactly this should look like is less clear. 70 percent of managers are in favor of developing AI in such a way that humans retain control and can intervene at any time. At the same time, 42% of employees believe that their company does not have a clear idea of which systems should be fully automated and where human intervention is appropriate.

Another AI survey conducted on behalf of Workday underlines concerns about the responsible and effective use of artificial intelligence in the company: In it, nearly three-quarters (72 percent) of executives say their organization lacks the necessary skills to fully leverage AI and ML. At 76 percent, the proportion of those who rate their own knowledge of AI and ML applications as needing to be improved is even slightly higher.

AI governance

When asked about a future in which artificial intelligence is part of everyday life, 42% of managers and 36% of employees cite framework structures and regulations within the company as the most important building blocks for trustworthy AI. The lack of transparency regarding AI regulations and guidelines in companies is perceived as problematic. Three out of four employees complain about the lack of cooperation from their employer when it comes to AI regulation; four out of five state that their company has not yet published any guidelines for the responsible use of AI.

AI undoubtedly offers immense opportunities for the transformation of the economy. However, our survey also shows that executives and their teams do not know exactly how their organization intends to use artificial intelligence in the workplace, nor do they have much confidence in this capability. To close this trust gap, companies need a comprehensive strategy for AI responsibility and governance. Transparent communication and convincing implementation are particularly important, as this is the only way to find the right balance between innovation and trust. Artificial intelligence and machine learning will only be successful if their use follows ethical standards. When gradually introducing AI-based technologies, we pay particular attention to involving customers and employees and gaining their trust. 

If ERP users want to derive maximum business value from AI and ML, they should first align expectations with reality and prioritize initiatives aimed at solving specific problems. This means choosing AI and ML options that match the business strategies and requirements of their own organizational structure and processes, and defining a clear roadmap with clear goals and metrics.

workday.com