As the Information Technology Officer at Neil Hoosier & Associates, Inc. (NHA), I’m excited to share insights into how small businesses like ours are embracing the opportunities presented by generative AI while carefully navigating the challenges. This journey has been rewarding, but it also comes with unique obstacles. Here’s a closer look at our experience, the hurdles we’ve faced, and the exciting potential for enhanced efficiency and innovation.

Getting Buy-In Across the Team

Introducing generative AI tools to any organization, especially one with a diverse range of experiences, brings a learning curve. Many of these valuable employees have honed their skills over decades and are understandably cautious about adopting AI tools.

We’re working to bridge this gap by hosting AI workgroup sessions that demonstrate how generative AI can simplify day-to-day tasks, ultimately improving their work experience. It’s about building trust in the technology and showcasing its potential to assist, not replace, their valued expertise.

Verifying Accuracy: An Ongoing Challenge

One of the most significant challenges we’ve encountered with generative AI tools is verifying the accuracy of the information they produce. AI-generated answers can be highly convincing but may contain subtle inaccuracies that can have major impacts on our work. In areas like Quality Assurance, Information Security, and Business Development, accuracy is paramount, so we’ve implemented layers of verification to corroborate AI outputs before they inform decision-making or are shared with clients. This has meant establishing clear processes for human oversight and prioritizing tools that allow us to track and verify their sources.

Choosing the Right Tool for the Right Role

With so many generative AI tools available—such as Microsoft Copilot, Perplexity, ChatGPT, among others—figuring out which tool is the best fit for each role has been a learning process. Here’s a look at how we’re approaching this:

• Quality Assurance & Testing: We have found that Perplexity excels for code review and ChatGPT for test case generation, both with a focus on minimizing errors and enhancing reliability.
• Outreach and Education: For our outreach initiatives, ChatGPT offers reliable, engaging and accessible content creation, though we prefer Perplexity for research purposes. Search results are more up-to-date and transparent about the way in which Perplexity “thinks”.
• Information Technology & Security: Where precision is critical, we need AI tools that provide vetted, reliable data that aligns with security best practices. These departments have found ChatGPT and Perplexity for coding assistance, troubleshooting, and policy drafting and updates. We’re also trialing tools that support both automated threat detection and enhanced user support.
• Contracting, Business Development, HR, Accounting, and Finance: Each of these roles requires unique considerations, whether it’s for drafting contracts, analyzing market trends, managing hiring processes, or processing financial data. ChatGPT has proven useful for report generation and data analysis. The IT team is working with these departments to build a custom Large Language Model (LLM) chatbot using Microsoft CoPilot Studio for employee policies and procedures. We’re experimenting with AI capabilities that align closely with each function’s unique needs, keeping an eye on compliance and accuracy.

The trial-and-error involved in matching each tool with its optimal use case has underscored that a one-size-fits-all approach doesn’t work when it comes to generative AI. Our commitment to continual assessment and employee feedback is helping us refine our tool choices as we go.

Measuring Efficiency Gains

One of the most exciting developments has been hearing from staff who report that they can now perform tasks faster and more effectively with the help of AI tools. We’re seeing benefits across multiple areas, from streamlined document generation to accelerated data analysis. However, quantifying these gains poses its own challenge. We’re actively working to develop metrics that capture improvements in speed, complexity, and accuracy. By analyzing these metrics, we hope to build a clear picture of AI’s impact on our productivity, empowering us to make data-driven decisions about future investments in AI technology.

Looking Forward

Our journey with generative AI is ongoing, and while it requires a thoughtful and strategic approach, the potential benefits are enormous. By investing time and resources to understand and optimize these tools, NHA is poised to stay at the forefront of technological advancement, enhancing both our internal operations and the service we provide to clients. We’re committed to navigating the complexities of generative AI thoughtfully and responsibly, ensuring that it serves as an asset to our team and aligns with our values.

I invite you to share your thoughts and experiences with AI in your organizations. Together, we can navigate this new landscape and unlock its full potential for small businesses everywhere.

Stay tuned for more updates as we continue to explore, learn, and grow in this exciting new era of AI-driven innovation!