CEOs play a big role in identifying the biggest gen AI opportunities—and making them happen.

ChatGPT’s extraordinary leap to fame has made generative AI the pop-culture celebrity of technology: It’s impossible for CEOs to ignore.

Unlike other technologies, generative AI’s potential impact on organizations is enterprise-wide. That makes it distinct from technologies like cloud, which is about productivity and cost optimization, and blockchain, which fits narrow use cases like improving visibility in the supply chain.

Generative AI is far more versatile. With its ability to convert text and images into new content, generative AI reaches across every corporate function and level. Its potential impact on strategy—and especially its ability to create value—puts it squarely within the sights of CEOs.

The challenge for CEOs is to get over the “fear of missing out” around generative AI and prioritize a careful, measured approach that identifies opportunities and lays the groundwork to make them happen.

Pragmatic next steps for CEOs

Here is a blueprint for how CEOs can lead their companies’ adoption of generative AI. The approach combines classic critical business thinking with tangible actions:

1. Identify key business priorities for generative AI. Despite the intense pressure on executives to do something—anything—with generative AI, the first key step is to define the problem to solve. CEOs can help narrow and refine their companies’ scope for using generative AI by emphasizing its potential as a tool in areas that can best capitalize on the technology’s facility with language and text.
   
   Possible areas include:
   
   • Documents and knowledge bases. Use generative AI to make it easier for employees to access data they already have access to. Human resources and the legal department are common targets for generative AI proofs of concept because both rely on a wealth of documents and knowledge bases. This makes these business functions a natural fit for large language models’ ability to automate the harvesting of information.
   
   • Customer service. Generative AI has brought the fail-fast approach back into action, and nowhere is that more apparent than in customer service, which is one of the technology’s most natural roles. For example, large language models (LLMs) trained on knowledge bases and other internal data can be used to enable customers to quickly find the information they’re looking for through conversational search.
   
   LLMs can also help with language translations and the creation of summaries for agents to use in their conversations regarding products and services or recommendations. The key is to get started. Don’t let perfect be the enemy of good.
   
   • Faster time to market for IT services. IT and engineering leaders are targeting productivity gains with generative AI—and a spirit of experimentation. There’s an emphasis on sandboxing as well as red-teaming generative AI-written and debugged code.
   
   In addition to helping data engineers better manage the flow of existing data and create new data to drive growth and aid decision-making, we see great potential for using LLMs for high-value tasks, such as data preparation and knowledge discovery. Businesses can also put LLMs to use as knowledge workers, asking them to reveal code weaknesses or interesting insights about a particular dataset.
   
   We also see companies focusing on how generative AI can improve security through its facility for quick, efficient threat analysis and the log reviews that make extensive use of text and language.
   
   
2.  Encourage collaboration in the C-suite—and beyond. The front lines of any generative AI agenda are crowded with executives and stakeholders. It’s the CEO’s job to ensure everyone works as a team, and that business executives are actively engaged with the data and technology leadership teams.
   
   Doing so often requires eliminating the corporate red tape that can hamper progress, such as establishing policies about data sharing and copyright protection, as well as guarding against turf wars. Often overlooked is governance, especially as companies begin to put a monetization value on generative AI.
   
   To shape their companies’ generative AI efforts, CEOs need to be mindful that scaling the technology will incur significant costs to build out the infrastructure and processing capabilities needed to churn through massive amounts of data.
   
   
3. Embrace the ecosystem. Simply put, generative AI is too complex for enterprises to figure out on their own. There are several ways for CEOs to begin building the network of partnerships they’ll need to capitalize on the technology.
   
   One is by elevating the conversation within the boardroom by bringing in technology partners for presentations. Another is to involve their company in external industry conversations through participation in industry councils, forums and panels.
   
   Don’t forget the internal-facing ecosystem–employees, board members, shareholders and value chain partners–where CXOs dominate and can advance the message of how valuable partnerships can be in shaping a successful generative AI agenda. In particular, the participation of employees is critical to increasing the innovation output from generative AI.

Generative AI reaches across every aspect of organizational activity. As such, it represents a unique opportunity for CEOs to help the business move beyond the generative AI hype with a strategy that lays the groundwork for realizing the technology’s potential.