Don't try to optimize your strongest departments with your first AI project. Instead, target 'layup roles'—areas where processes are broken or work isn't getting done. The bar for success is lower, making it easier to get a quick, impactful win.

The path to adopting AI is not subscribing to a suite of tools, which leads to 'AI overwhelm' or apathy. Instead, identify a single, specific micro-problem within your business. Then, research and apply the AI solution best suited to solve only that problem before expanding, ensuring tangible ROI and preventing burnout.

To introduce AI into a high-risk environment like legal tech, begin with tasks that don't involve sensitive data, such as automating marketing copy. This approach proves AI's value and builds internal trust, paving the way for future, higher-stakes applications like reviewing client documents.

Avoid deploying AI directly into a fully autonomous role for critical applications. Instead, begin with a human-in-the-loop, advisory function. Only after the system has proven its reliability in a real-world environment should its autonomy be gradually increased, moving from supervised to unsupervised operation.

To mitigate risks like AI hallucinations and high operational costs, enterprises should first deploy new AI tools internally to support human agents. This "agent-assist" model allows for monitoring, testing, and refinement in a controlled environment before exposing the technology directly to customers.

To navigate the high stakes of public sector AI, classify initiatives into low, medium, and high risk. Begin with 'low-hanging fruit' like automating internal backend processes that don't directly face the public. This builds momentum and internal trust before tackling high-risk, citizen-facing applications.

For its complex payroll product, Shure isn't attempting full automation on day one. It's taking a piecemeal approach, starting with one country (Nigeria) and keeping humans in the loop. This allows them to refine AI agents in a controlled environment before scaling globally.

To bridge the AI skill gap, avoid building a perfect, complex system. Instead, pick a single, core business workflow (e.g., pre-call guest research) and build a simple automation. Iterating on this small, practical application is the most effective way to learn, even if the initial output is underwhelming.

For companies wondering where to start with AI, target the most labor-intensive, process-driven functions. Customer support is an ideal starting point, as AI can handle repetitive tasks, leading to lower costs, faster response times, and an improved customer experience while freeing up human agents for more complex issues.