CIO Fireside Chat Recap: Leveraging Generative AI in the Microsoft Cloud

A recent survey by Deloitte found that businesses are embracing generative AI, with 92% of Fortune 500 companies already adopting some form of AI—especially among Millennial and Gen Z employees. Microsoft kicked off their “Microsoft 365 Copilot Wave 2” announcement event noting that almost 60% of those Fortune 500 companies now use Copilot AI. In light of the industry buzz around generative AI, Blue Mantis hosted a thought-provoking discussion during our September 2024 CIO Fireside Chat webinar, featuring John Treadway, CEO and Co-Founder of AI Technology Partners (AITP). Our conversation ventured into the transformative realm of generative AI within the Microsoft cloud ecosystem, and I’m excited to share some insights from the live CIO Fireside Chat with you:

How to Adopt AI at Your Business

John and I started our discussion outlining the journey currently being taken by today’s early adopters of generative AI, emphasizing the critical role of strategic planning for the future. John Treadway brought an interesting view of this journey towards broad generative AI adoption in business. “It’s about education and understanding,” John noted. From his perspective as both a business executive and technologist, John noted that IT leaders should avoid overthinking it and “moving forward gives you clarity.” There is a necessity for organizations to embrace AI technologies proactively and partnering with consultants if your IT team doesn’t have in-house AI expertise and you want to avoid “analysis paralysis.”

Most importantly, John emphasized the importance of CIOs and CISOs working together to assess cybersecurity needs before deploying any AI tools. For example, a company deploying Microsoft 365 Copilot will need to expose their internal business data and files to the Microsoft Graph API. Data from the Graph API feeds into the AI that powers the Microsoft 365 Copilot assistant to provide customized results from employee questions. That’s why it’s crucial for all company data to be properly secured and access via the AI is appropriately managed to mitigate risks. Additionally, John underscored the significance of training and enablement for users to effectively utilize Microsoft 365 Copilot, suggesting that without proper training, the adoption and value derived from Copilot will be extremely limited.

How Much Should a Businesses Spend on AI?

For the CIO or CISO developing an AI strategy, a key decision point is the cost of deploying AI. The oft-cited example is Microsoft 365 Copilot. An add-on for existing Microsoft 365 subscription plans, the Copilot AI assistant costs an additional $30 per month, per user. Current pricing for the Microsoft 365 plan specifically for small businesses (fewer than 300 employees) is $22 per month, per user. For the lowest cost enterprise plan (E3, including Microsoft Teams for a true apples-to-apples comparison of plans), businesses pay just under $40 per month, per user. That’s why the CIO who’s asking a CEO or CFO for an extra $360 per user annually needs to have a firm grip on Microsoft 365 Copilot’s return on investment.

“You need to understand what the impact is on your data estate before you go all in with a Microsoft 365 Copilot implementation,” John explained. He gave some examples of how generative AI is revolutionizing user interfaces and experiences, significantly enhancing productivity for his own business. John advocated for IT leaders to look for opportunities to enable the entire company with generic AI capabilities while also investing in high-value custom AI solutions for specific business functions. One example could be to look at cost savings from enhanced operational efficiency, suggesting that businesses can start with smaller, proof-of-value projects to assess the benefits before committing to larger investments. This approach helps CIOs to maximize their ROI for an AI deployment.

John addressed the concern about the “massive capital outlay” for using generative AI by differentiating between the costs of building core generative AI capabilities and applying generative AI solutions. He explained that while training large language models (LLMs) like GPT-5 are projected to cost up to a billion dollars, those investments are made by Microsoft, xAI, and other large companies to actually build the LLM to productize a generative AI solution. However, applying generative AI in a business context can be much more affordable. John mentioned that for smaller projects or applications, the costs could range from tens to a few hundred thousand dollars, depending on the complexity and scope of the solution. He emphasized that the investment in generative AI should be aligned with the potential return on investment.

How to Select and Budget for AI Tools at Your Business

John addressed an audience member’s question about selecting from multiple generative AI tools by emphasizing the importance of understanding the specific needs and context of one’s organization. He suggested considering the integration capabilities of each tool with existing systems, like Microsoft 365, Salesforce, and AWS, and assessing the cost implications relative to the budget for 2025. John recommended focusing on one tool that best aligns with the company’s operational needs and budget constraints, rather than investing in multiple tools. He highlighted the need for a strategic approach to adopting generative AI tools, ensuring they complement the company’s existing technology stack and support its operational goals.

Another attendee asked John about potential usage scenarios for their small (under 500 employees) manufacturing firm to take advantage of generative AI. John emphasized the importance of understanding the specific context of a company’s operations to identify how generative AI can enhance operational efficiency. He suggested that generative AI could be applied in the manufacturing context to categorize complex data, such as customer feedback or operational metrics, to improve decision-making and streamline processes. John then highlighted the potential for generative AI to assist in creating more efficient workflows and enhancing customer service by quickly analyzing and providing automated responses to customer inquiries or feedback. “We’ve done this for other clients,” John said and pointed out that Blue Mantis and AITP helps CIOs think through the process of determining ROI for AI deployments based on desired business outcomes.

Blue Mantis Helps CIOs Successfully Deploy AI

As we navigate this new era of technology, the time to act is now. “You need to have that Board-level conversation about AI,” John said, “and if you’re having trouble getting the business case and figuring out where the real value is, that’s what we do.”

If you’d like to have a conversation with a Blue Mantis AI expert and our partners at John Treadway’s AITP, then connect with us. Let’s meet to explore how we can assist you in integrating Microsoft Copilot 365 and custom-designed enterprise AI solutions into your business strategy.