ments to creating dashboards using plain language. For example, “Which five inspec- tors cited handwashing violations in routine inspection of retail food establishments in the past year,” is a request that can now be met in some systems. It is just going to get better and better. When it comes to great ideas about AI, let me introduce a decision-making model known as a feasibility matrix (Table 1). In this model, your AI adoption commit- tee agrees on a rubric and assigns scores to those measures. Common measures might be mission alignment (e.g., does the ini- tiative impact public health), increased e ciency, and nonfinancial benefits (e.g., customer satisfaction), followed by poten- tial headwinds such as technical feasibility, internal readiness, and external readiness. Internal and external readiness might, for example, consider compatibility with your AI policy. Each score need not be completely scien- tific and will change over time. Readiness and technical feasibility can change every day. Reviewing and updating your feasibil- ity matrix should be a quarterly exercise or part of your strategic planning. Chat- GPT4 is advertised to be much better than ChatGPT3. Also, many organizations are regularly releasing their versions, such as LLAMA2 from Meta, Bard from Google, and Claude from Anthropic. In Table 1, one might agree that “meeting deliverable summaries” and “education and training” are viable projects right now. What Is on the Horizon? It is no longer controversial to suggest that we will have personal AI assistants, auto- mated responses to external queries, and software systems that talk to each other using just natural language. If you are in the work- force 10 years from now, you will be among the leaders who frame and usher in many of these changes. For environmental health professionals, there will be massive changes not only in regulator o ces but also in the kitchens and o ces of the establishments you regulate. Corresponding Author: Darryl Booth, General Manager, Environmental Health, Accela, 2633 Camino Ramon #500, San Ramon, CA 94583. E-mail: dbooth@accela.com.
FIGURE 1
Decision Chart to Determine If It Is Safe to Use ChatGPT
Adapted from The Prompt Index (https://www.reddit.com/r/ChatGPT/comments/16ailop/is_it_safe_to_use_chatgpt/ ?rdt=34544).
TABLE 1
Sample Feasibility Matrix
Mission Alignment
Increased Efficiency
Nonfinancial Benefits
Technical Feasibility
Internal Readiness
External Readiness
Draft inspection reports Respond to public inquiries
2
2
1
1
1
1
2
2
2
2
2
2
Education and training
3
2
2
2
2
2
Natural language dashboards Regulatory assistant Meeting deliverable summaries
2
2
1
1
1
1
1 2
2 3
1 1
1 3
1 2
1 3
Note. Ratings are based on a 5-point Likert scale (0 = not feasible and 5 = highly feasible).
expectations. After all, if Word will help me write a determination letter, why wouldn’t my inspection software do the same?
Furthermore, your inspection and permit tracking software will be able to do so much more, from helping with inspection com-
35
November 2023 • Journal of Environmental Health
Powered by FlippingBook