AI can be complicated, with many different solutions, tools, and platforms available. To make the most of GenAI, you first need to understand the basics and set a solid foundation with your leadership team.

GenAI leverages multiple models to create new content and solve problems. These models — like GPT-4 for text, VALL-E for audio, and DALL-E for images — rely on a wide variety of data for training and inferencing. This data can impact the safety, security, and cost of a solution. For instance, using proprietary data for training can enhance the accuracy of outputs, but may put sensitive information at risk.

That’s why it’s important to have the right implementation partner so that the nuances of these kinds of decisions are considered and enforced.

Each model has unique capabilities, like text, voice, image, or a combination of these, that represent human-like interactions. Understanding these capabilities is essential for an effective GenAI strategy. When approaching leadership, you need to be clear about which model you intend to focus on and how it can be used.

To make GenAI more appealing to leadership, it’s important to highlight clear use cases within your business.

Once you’ve outlined relevant use cases, dive deep into what can be expected. Describe how the model will work, the areas of the organization it will touch, who on the team will be impacted, and the estimated cost of implementation (more on that in the next section).

Finally, share compelling, real-world case studies that paint a clear picture of the impact of GenAI. By demonstrating tangible examples, you’ll make the advantages more relatable and easier to grasp.

Here are some examples:

• Customer support: Faster response times and more accurate information.
• Product recommendations: Increased sales and customer satisfaction.
• Enterprise search: More efficient information retrieval and decision-making.
• Content creation: Saves time and generates higher-quality content.

Be realistic about the hurdles your organization could face during GenAI implementation. Identify, understand and be transparent about risks. You don’t want to sell your GenAI solution to leadership, only to have them be blindsided by the unexpected.

Here are a few common challenges your team needs to be aware of:

GenAI involves upfront costs for training, deployment, and scaling up and out. Use simple terms to explain the pricing model and focus on the long-term ROI.

Many generative AI prediction services are priced using token-based pricing. This means costs are based on the number of input and output tokens (words or phrases) used in your training inputs. Tokens are a unit of text that an AI model processes, and generally one thousand tokens are about 750 words.

Highlight the need for an initial investment for retraining the model to your data — that means updating the AI systems to recognize and adapt to your specific business information and needs. This can be expensive but will provide more effective prompting and results. The solution needs to be accurate to achieve the desired results.

One way to efficiently accomplish this is to consider introducing tools like retrieval-augmented generation (RAG), which can optimize the output of GenAI models. It offers a cost-effective solution to inject new information into a model without the need for extensive retraining. RAG can be ideal for businesses that work with frequently changing or proprietary data, as it aims to protect data, increase reliability, and offer fact-checked outputs.

When building the case for decision-makers, make sure you understand the cost model you’re proposing. Clearly explain the pricing structure and the necessity of the initial investment so they view GenAI as a strategic investment rather than a barrier. Emphasize the long-term return on investment by showcasing how these initial costs will lead to significant improvements in efficiency, productivity, and overall performance.

Data security is essential. Emphasize the importance of setting proper permissions and safeguarding sensitive information to prevent breaches.

Existing generative AI models have often been trained on vast amounts of data. Hallucination, bias, security, and safety challenges are typical when integrating GenAI. Leadership should be aware of these issues, but also that there are frameworks to mitigate them.

When you’re presenting your case, make it clear you have thoroughly considered data security and the importance of using clean data. Explain how these measures are imperative for preventing breaches and ensuring the AI system operates fairly and effectively.

The business world is fast-paced, with many of the existing technology roadmaps now outdated. This new era of tech calls for rapid experimentation and proofs of concept to quickly validate assumptions and confirm ROI.

To communicate this in your pitch, focus on areas where you can quickly demonstrate value and stress the importance of a committed approach to GenAI adoption. The best way to do this is to go deep into one area, instead of spreading yourself too thin by trying to solve a little of everything at once.

Make sure your organization is ready to integrate these tools. For instance, there may be talent gaps or areas that need improving before you move forward.

When you approach your leadership, you need to be able to ensure that there will be as smooth a transition as possible. Don’t overpromise, as realistically there may be hurdles to overcome (and you should identify them), but as part of your business case, you need to showcase how you plan to overcome them.

Building a business case for GenAI involves more than just understanding the technology. It’s about translating its potential into real business benefits and addressing the practical challenges head-on.

These tools have the potential to revolutionize your business operations, making processes more efficient and customer experiences more personalized. But it’s critical these points be communicated effectively to your leadership team, address their concerns and highlight tangible results.

There is considerable upside to adopting GenAI, but this work cannot be ad hoc or done off the corner of someone’s desk. Its impacts are significant and far-reaching, requiring strong leadership and strategic focus. 

Nominate an executive sponsor who can champion the adoption of GenAI within the organization. Identify those who are familiar with the topic or are eager to have a voice in the discussion. This working group will be responsible for identifying hurdles, roadblocks and areas of the business where change may cause friction. It can also reduce the challenges of GenAI adoption through positive training, proof-of-concepts, hackathons, and highlighting areas where competition is adopting GenAI.

By taking these steps, you will see far greater success in making AI more accessible and engraining it into your organization’s culture at an early stage.

To build a compelling business case for GenAI, make sure you focus on its transformative potential and practical benefits. And when pitching, please remember these four things:

1. Highlight how it improves efficiency, customer service, and innovation
2. Explain the cost model and emphasize the need for initial investments in retraining the tool with your organization’s data
3. Address data security and clean data use to mitigate breaches
4. Communicate these points clearly to decision-makers, showcasing long-term return on investment and addressing any potential hurdles.