GenAI Do's and Dont's

In today's fast-paced business environment, enterprises are increasingly turning to Generative AI (GenAI) to streamline operations and gain a competitive edge. This transformation is not just about automation; it's about reimagining how businesses operate in the digital age.

The Challenge of Legacy Processes

Many enterprises struggle with inefficient legacy processes that hinder productivity and innovation. These outdated processes, supported by outdated and inflexible systems often lead to:

• Data silos and poor information flow

• Time-consuming and repetitive manual tasks

• Inconsistent decision-making processes

• Difficulty in standardisation and scale operations

The major challenge revolves around the understanding of what the problem is. Operational teams know what's wrong and they don't like doing, technical teams think they understand what those issues are, but are operating in a constrained (budget, talent and variety of options) circumstance, and make the best with what they have, and leadership rarely understand why "cutting 10%" on spend might have catastrophic consequences.

In many past experiences, the first step is getting everyone in a room and agree on what things mean, whose responsibility it is to fix it, and how can impact be measured. It's not a single person's job, its a multidisciplinary team effort, and even a company-wide one.

GenAI: The Catalyst for Operational Transformation

GenAI is proving to be a game-changer in addressing these challenges. By following a user centric process to map end to end processes, and breaking them down into components (user action, output, systems used, data and source), teams can then evaluate every step of the process and decide what's the best choice to optimise or completely reinvent it.

Generative AI is not a silver bullet that with solve all your problems, but it does give the equivalent of super powers. Having said that, it's important to validate whether it is the right tool for each of those steps - so make sure you start with this list:

Patterns to apply Generative AI	Business challenge example
Bulk upload and processing	Read and extract information from 100’s of similar documents
Summarization	I need a 1-minute update about X
Version generation	I need to inform different stakeholders about the same event
Interpretation	I need to review and find discrepancies and opportunities to improve
Outputs review	I need to understand what type of document this is, its quality / completeness
Copy / paste across systems	I need to gather information from several systems and bring it together
Training / best practice	I need to know what’s the logic applied and whether I’ve missed something

With these opportunities to generate efficiencies, companies of any size can take advantage of these tools, as long as they are thinking about how it needs to be used, for what purpose and by whom.

It is important to follow a very thorough process to make sure what you deliver a scalable and coherent outcomes. And there are some elements to understand as temporary crutches to ensure the solutions are fully usable across the business. The way you design the lifecycle is going to be paramount to get scalable and relevant outcomes.

Real-World Impact

JPMorgan Chase's implementation of AI-powered contract analysis showcases the transformative power of GenAI in operations. The bank reduced document review time from 360,000 hours to mere seconds, dramatically improving efficiency and accuracy.

Integration Challenges and Solutions

While the benefits are clear, integrating GenAI into existing enterprise systems can be challenging. Key considerations include:

• Ensuring data quality and compatibility

• Addressing privacy and security concerns

• Training staff to work alongside AI systems

• Aligning AI initiatives with overall business strategy

  
  

As GenAI continues to evolve, its potential to revolutionise enterprise operations grows exponentially. By embracing this technology, businesses can transform chaos into clarity, driving efficiency, innovation, and growth in an increasingly competitive landscape.

We will cover pitfalls with some horror stories soon, in the meantime, feel free to share what has been either making a difference to be successful, or has ground your AI projects to a halt.

References

1. Daugherty, P. R., & Wilson, H. J. (2018). Human + Machine: Reimagining Work in the Age of AI. Harvard Business Review Press.

2. The AI Alignment Podcast. (2023). "AI in Enterprise Operations." [Podcast]

3. Harvard Business Review. (2023). "Operational AI: The Next Frontier for Enterprise Transformation." IdeaCast. [Podcast]