Can We Even Do AI?
Artificial Intelligence (AI) is everywhere. Successful early adopters share a common trait: a thoughtful approach to how they prepare. Previously, we discussed the foundational knowledge needed before jumping into AI. Now, let's assess whether AI is the right fit for your organization before committing significant resources and potential disruption.
theBigRocks of Change(c) Model for Introducing & Adopting Artificial Intelligence
2. Proof of Concept: Test & Validate + Establish AI Governance
Once your organization has enough knowledge to determine AI is probably worth looking into, the second phase of your adoption journey should be focused on validating how ready your business and technology foundation is to support it.
We recommend a methodical “Proof of Concept” approach where you’ll:
• Select a Targeted Business Challenge: Identify a specific process area and sub-unit within your organization where AI can address a significant issue or improve efficiency. (For example: reducing time a given group spends in meetings) Choose a challenge that’s both impactful and measurable. We highly recommend you choose a team that already looks forward to leveraging AI. (They’ll be more apt to tolerate potential bumps in the road.) Identify AI Change Champions from within the area where the challenge will be addressed.
• Implement a Small-Scale Proof of Concept: Develop and deploy a limited, controlled AI solution to test its feasibility and effectiveness. Ideally this group should include no more than a dozen users. If you have not chosen your preferred AI solution yet, you may want to run simultaneous Proof of Concept rounds with more than one tool and compare results to finalize that choice. Determine a specific list of business scenarios the Proof of Concept will engage and set a short time window. Avoid trying to measure too many scenarios - we recommend fewer than 10. There will be plenty of time to add complexity later - this stage is mostly meant to prove you can make the solution work within a willing team.
• Build Your Tech Team & Governance: Ensure your technical team has a chance to build their configuration skills and feels comfortable supporting users. Establish your initial AI Governance Model. (For example: Identify who can do what AI tasks with what data and what data needs to blocked or be subject to limited access.) Be ready to update your AI governance based on the lessons you learn as you move forward.
• Gather Feedback and Refine the Solution: Actively involve your AI Change Champions & end-users in the Proof of Concept process. Collect their feedback to identify surprises, gaps or issues. Gather potential future scenarios that they think might be worth trying in the next round. Use this valuable input to refine and optimize the AI solution before considering a broader rollout.
• Build the Business Case: As the Proof of Concept starts to generate momentum, take the time to translate the numbers you’re collecting into a broader business case for rolling the tools out further. Define your initial budgeting needs for things like licenses, support and training. You can leverage dozens of great online sources to calculate potential Return on Investment (ROI) based on your organization’s industry vertical and typical business roles. One of our favorite resources for this part of building the business case is the site JobsGPT by SmarterX.ai. It’s free and includes a great AI benefits calculator. Just be sure to review any results you get to make sure they’re valid for your company, your market, and your actual business roles.
In our next article, we’ll start to prepare for the broader rollout of AI tools to your organization by conducting a formal “Pilot Phase” that extends your gains into parts of the organization that may not have asked for AI.
-Steve
Reach out to theBigRocks if your team is considering a rollout of artificial intelligence tools like ChatGPT or Microsoft Copilot(c). We can help you walk through this model and plan for success.
