#GenerativeAI has revolutionized computing, offering remarkable capabilities. However, it’s important to recognize that its true value lies in task-based automation. While generative AI has its merits, I remain skeptical about it becoming the primary driver of business processes in the near future. Ultimately, it is the human-led processes that are the backbone of any organization.

Complexity of Business Processes

Business processes are intricate, involving multiple stakeholders, extended timeframes, and often branching and looping actions. One well-known example is the quote-to-cash process, which encompasses the entire journey from initial customer interaction to payment collection. Automating such a complex process, like quote-to-cash, can be incredibly challenging. Some businesses have achieved success by automating specific steps based on transaction monitoring and system integration. This is known as task-based #automation .

Human Ownership of Business Processes

Considering the future of AI, productivity, and labor, it is unlikely that humans will readily relinquish control over business processes. There are two primary reasons for this. Firstly, AI’s decision-making track record falls short of perfection, with hallucinations, occasional errors and a lack of transparency in how responses are generated. Hence, it would be unwise to exclude human intervention in the form of human-in-the-loop automation. Trust in AI’s decision-making remains a crucial factor.

The Role of Humans in Current System Architectures

Secondly, humans are deeply involved in designing and organizing enterprise systems. These systems are typically designed in accordance with Conway’s Law, which states that organizations create systems mirroring their communication structure. Current system architectures are not optimized for operation by digital entities but rather tailored for human decision-makers. This includes considerations of data capture, organization, identification of related data across systems and departments, and the concept of data gravity—the term coined by Dave McCrory to identify the phenomenon that services and applications tend to accumulate near data due to the data’s mass, thus increasing the time and costs to move, rather than be optimally placed to maximize business performance and operations.

Generative AI and Quote-to-Cash Process

While it is possible to provide sufficient data to a generative AI to develop a complete business process, such as quote-to-cash, I believe there are challenges to overcome. A generative AI might successfully create a process for a single system like Oracle or SAP. However, it would likely be biased towards existing human practices, resulting in a garbage-in, garbage-out scenario. Furthermore, there is a risk of unintentional missteps that could negatively impact customer interactions, such as focusing on legalities without the finesse and empathy of a human account executive. Moreover, these complex business processes often involve more than just one system that needs to be updated to ensure proper completion

Conclusion

Generative AI has incredible potential, particularly in task-based automation. However, it is important to recognize the complexity of business processes and the significant role humans play in their design and execution. While generative AI may assist in certain aspects, it is unlikely to replace humans as the primary operators of business processes. Striking a balance between human expertise and AI automation will continue to be crucial for organizations seeking to optimize their operations.