Module 1
AI Governance: Creating Trust, Compliance, and Data Privacy

Module 1
The Future of AI in Business

Module 1
Glossary of Common AI Terms

Getting High- (and Higher-) Quality Results

Achieving high-quality results with prompt engineering is iterative and requires a combination of clear problem definition, experimentation, context provision, and sometimes advanced techniques like fine-tuning.

Below is our step-by-step approach to achieving high-quality results with AI.

1. Define the Goal – The process involves moving from point A to point B with AI as an accelerator. Determining point B (the desired outcome) is crucial.


2. Define the Problem – Clear problem definition is crucial. If unclear, take a break, reflect, and use tools like ChatGPT to help structure your thoughts.


3. Use the Scientific Method – Develop a hypothesis, conduct experiments, analyze the results, and incorporate findings for continuous improvement.


4. Utilize Prompt Iteration – Assess results against a pre-defined threshold, applying constraints and leveraging AI to solve the problem quickly. Iterate based on outcomes. If unsuccessful after several tries, break the problem into smaller parts.


5. Quality Assessment – If results are ≥50% of the desired quality and save significant time, consider refining the prompt further. If results are <50% of the desired quality, break the problem down further.

Best Practices

  • Use “zero-shot” prompts (no examples) to test the water.
  • Provide gold-standard examples to boost accuracy.
  • Incorporate more context either in prompts or through conversations with AI.
  • Consider fine-tuning by adjusting the model based on a set of examples, enhancing its predictions.

Similar to training a human, by providing instructions, showing exemplary work, and giving context, over time, mastery is achieved.