How your organization can capitalize on generative AI

Published On: April 24th, 20244 min read

What do an art gallery, a global automobile manufacturer, a power management company, and a government agency have in common? They are all looking for ways to use generative AI (GenAI) to deliver value, improve customer experiences (CX), or increase productivity. 

According to Gartner, by 2026 more than 80% of enterprises will have used GenAI application programming interfaces (APIs) or models and/or deployed GenAI-enabled applications.  

Many business and technology leaders are facing similar questions as they consider implementing GenAI at their organizations. Despite some uncertainty, according to Forrester, 71% of enterprises are already experimenting with actual use cases for GenAI, representing one of the fastest mass adoption rates of a new technology in the enterprise. Moreover, Forrester predicts firms that actively harness GenAI to enhance experiences, offerings, and productivity will realize outsized growth and outpace their competition.  

Nearly every week, new GenAI tools, plugins, and solutions enter the marketplace. Interestingly, a Microsoft-commissioned IDC study of 2,109 enterprise organizations found that organizations realize a return on their AI investments within 14 months. For every $1 a company invests in AI, it realizes an average return of $3.5. However, it can be challenging to identify the best possible investments. 

One way to cut through the hype and construct a strategy to evaluate GenAI-based solutions is to focus on how your organization can drive business value. Using a crawl-walk-run approach to get started with GenAI is also advantageous to ensure your investments are achieving the desired outcomes. It’s also imperative to assess privacy, confidentiality, and liability issues before deploying GenAI at your organization. 

Some of the key areas to consider when getting started with GenAI include: 

  • Privacy: GenAI’s powerful language modeling and visual analysis features make it a powerful tool for summarization, content creation, and image generation. When interacting with GenAI, users must often provide contextual data such as documents, images or other media. The combination of contextual data and a question is known as a prompt, and these prompts may contain proprietary or confidential information not intended for external sharing. When using a GenAI service, the information in a prompt may leak outside the organization. Organizations can invest in user training and establish clear user policies to bolster security and confidentiality. Additionally, organizations should consider establishing firewalls to limit and manage access to data. Some larger GenAI and cloud platform providers allow organizations to run GenAI systems in a secure private cloud environment. Organizations must pay particular attention to managing privacy risks when using GenAI. 
  • Costs: Some activities around creating GenAI solutions, such as private GPTs and workforce training, are compute-intensive and require expensive hardware like GPUs. These costs can be especially high during pilot initiatives if they are not managed or controlled. IDC predicts that by 2027, spending on generative AI solutions, which includes software, related infrastructure hardware, and the IT or business services needed to implement the GenAI, will reach $143 billion. To better manage the costs of a GenAI implementation, it’s essential to strategically plan the project and define all processes, metrics and timelines before kicking off the initiative. In addition to traditional software project management metrics, monitoring metrics related to model selection, such as the number of parameters or size of the context window, is beneficial. 
  • Benefits: One benefit of tracking implementation metrics is the ability to quantify a GenAI solution’s impact. Continuous measurement helps monitor progress and provides valuable insight for prioritizing and optimizing future GenAI investments. These metrics include adoption rate, frequency of use, duration of sessions, model accuracy and error rate, and tracking harmful categories and topics. To maximize the benefits of a GenAI implementation, organizations should track:  
    • Business impact: Utilize GenAI to solve a specific business problem or achieve a particular outcome. 
    • Adoption measurements: Focus on user behavior and satisfaction.  
    • Quality measurements: Monitor whether the system is meeting performance goals, like speed and latency, and qualitative goals, such as accuracy. 
    • Implementation approach: Prioritize an incremental approach to using GenAI in the organization. 
    • Best practices: Collaborate with partners and peers to learn from their experiences. 
  • Time and Skills: The field of GenAI is still new, and there is a high demand for associated skills. While many organizations are eager to implement GenAI, taking measured steps, carefully planning the implementation, and bringing in the right skills at the right time is advisable. Defining an incremental approach with phased pilots and continuous operational and quality metrics tracking will allow the team to do fast prototyping and efficient innovation. In parallel, organizations should invest in training their existing workforce to adapt to the new technology and learn how to work with GenAI. Employees with legacy knowledge of an organization’s business processes who can also engineer solutions based on GenAI can be powerful differentiators.

Many smaller businesses that are new to GenAI and have budget constraints are seeking partners to help advise them on the best path forward. In fact, 52% of enterprise organizations report that a lack of skilled workers is their most significant barrier to implementing and scaling GenAI. It takes a village, as they say, and a good guide will walk with you at your pace and help you mitigate your organization’s unique risks. The ideal partner will focus on how your organization can drive business value with GenAI, design a tailored crawl-walk-run approach, and help assess any privacy, confidentiality or liability issues. 

For more information, visit: