Mar 28, 2025

getty AI is reshaping how we work, and our role as COO is evolving along with it. At Vultr, I’ve seen firsthand how AI models like generative pre-trained transformers (GPTs) can drive efficiency across the enterprise. But successful deployment isn’t just about adopting the latest technology—it requires strong governance, trust and effective change management. If you’re a COO looking to empower your employees with a GPT, here’s a straightforward roadmap for success. Form an internal AI committee. Many COOs at AI-mature enterprises emphasize the need for an AI Center of Excellence (CoE). But if a CoE sounds overwhelming (and I get it—it does), think of it instead as forming an internal AI committee. This cross-functional team—including your CIO, CTO, VP of Engineering and other leaders—champions AI adoption while ensuring governance, security and compliance. The committee’s role is to define policies for safe and responsible AI use, establish data governance that balances speed with trust and deploy an internal GPT client that genuinely empowers your employees. Just as importantly, they set the tone for AI adoption. Three-quarters of business leaders say quick AI adoption will help them maintain a competitive advantage. Employees are 1.5 times more likely to use AI at work —and nearly three times more optimistic about its potential—when their leaders demonstrate confidence and enthusiasm about the tools. Establish trust, safety and transparency (aka, a "no judgment zone"). Trust is essential when rolling out AI tools. Privacy and security are non-negotiable, but so is creating an environment where employees feel comfortable experimenting with AI without fear of judgment. Treat AI like any other productivity tool—something employees can explore freely to enhance their work. They shouldn’t be concerned that their AI interactions might be monitored. To build this trust—or as I call it, the "no judgment zone"—set a clear policy that AI interactions will never be under surveillance or used for performance reviews. At Vultr, the essence of our "Responsible Use of AI Policy" is simple: the principle of least privilege, be responsible with company data, you’re accountable for your work product and AI interactions aren’t tied to employee assessments. Leadership reinforces this message, ensuring our teams feel safe asking questions and contributing to AI improvements. This removes the "Big Brother" fear and creates a supportive culture of positivity around AI adoption. Adopt a layered approach to data privacy. Securing your GenAI requires the same diligence as any other enterprise system. Role-based access controls (RBAC) protect sensitive data while allowing employees access to relevant insights. A tiered approach—like a wedding cake—balances security and usability: • Outer Layers: Start with the least sensitive data, such as content on platforms like Confluence or intranets. • Middle Layers: Gradually move to slightly more sensitive data, such as customer support systems or IT ticketing platforms. • Inner Layers: Eventually, extend to more privileged data, like financial information, ensuring tighter controls at this stage. • Top Layer: Highly sensitive personal information should remain off-limits to GenAI applications. Pursue high-impact areas and show early wins. Identify your organization’s biggest pain points and start by applying AI where it delivers immediate value. Focus on processes bogged down by repetitive, time-consuming tasks. Areas like customer support, recruitment, sales and marketing offer big potential. A recent study by 451 , in partnership with Vultr, found that 84% of respondents saw improved customer service performance after deploying AI, while 89% reported better results in marketing (pg. 6). At Vultr, one of our biggest pain points was the time our support team spent searching through documentation to assist customers. To address this, we developed a self-hosted GPT client that indexes internal knowledge and customer interaction history, providing real-time, on-demand insights to our support team. The result? Faster issue resolution times, increased efficiency and high customer satisfaction. By tackling high-impact areas first, you can achieve quick wins to build excitement and trust before scaling across other areas. Build or buy? Assess your options. Deciding whether to build your own internal GPT tools or leverage external solutions depends on your organization’s technical capabilities and AI maturity. One of the advantages of being a COO for a tech-first company is having extensive resources at my fingertips. Given our expertise in AI infrastructure, it made sense for us to develop a private, self-hosted GPT solution tailored to our needs and integrated into our control plane, ensuring full alignment with our existing governance and security frameworks. For companies with fewer tech resources or those earlier in their AI journey, building and hosting a custom solution can be daunting. Nearly 30% of companies lack the infrastructure and talent to deploy AI effectively, and 21% cite budget concerns (pg. 14). In these situations, partnering with a trusted provider is often the best approach—letting you treat GenAI as an operational expense while relying on an IaaS provider that aligns with your data governance and security standards. This lets you maintain control and keep costs in check without compromising security or performance. Learn from early AI leaders. AI is evolving rapidly. Staying ahead of emerging trends often means looking beyond your organization for inspiration. Two companies that I find particularly innovative in their use of GenAI are Goldman Sachs and Walmart. Goldman Sachs is empowering its developers with a centralized AI platform that offers access to a range of large language models. This platform enables the rapid creation of secure and compliant AI applications, while their focus on code generation tools has boosted developer efficiency by about 20%. Walmart , on the other hand, is equipping over 50,000 corporate employees with GenAI tools such as "My Assistant" to streamline tasks like drafting and summarizing. Additionally, Walmart’s developer-focused tools are revolutionizing the coding and software deployment processes, driving greater efficiency and innovation across the organization. Crafting a culture of success. Whether you build in-house or go with an experienced AI infrastructure partner, success depends on more than just technology—it’s about creating a culture where AI use is encouraged and strategically integrated. By starting with a strong committee focused on building trust, you can lay the foundation for an impactful AI-first culture within your organization.