Context is King: Your Key to Better AI Results
Providing context is the best thing you can do to improve your generative AI (GAI) results. AI chatbots don’t know you and they don’t understand the context in which you’re working … unless you tell them. ChatGPT has some memory, but chatbots are masters of the middle. They just make assumptions about context based on their training data, which can result in generic results and excessive iteration to get what you really want.
The "Aha!" Moment: Context Tags in Action
This was driven home to me when I was working on a major project with Lex.page, one of my favorite AI tools. (Lex is a low distraction, AI-enabled writing environment. You can learn more about it here.) Lex recently added something called context tags that let you store specific contextual information such as the type of document, target audience, goal, tone and the like. Here’s the context information for my AI Goes to College context tag.
When you use a specific context tag, Lex’s AI chatbot will consider that context in its responses—the responses are tailored to the context. This is insanely useful since I use Lex for different kinds of writing tasks. I have other context tags for a specific journal article, an NSF grant proposal, and my doctoral seminar. As I continue to expand my uses of Lex, I’ll add more tags.
Proof in the Pudding: Context Makes a Difference
Using Lex’s context tags reminded me of just how important context is when using generative AI. If you want to level up your use of GAI, start adding context to your prompts. You’ll find that your responses are MUCH better, with minimal effort.
Here’s an example. I asked Claude the following:
Design a set of learning objectives for a class session on generative AI ethics.
Claude responded with a perfectly reasonable set of learning objectives, but the list was generic. There was nothing tailored to the class I teach.
Results were much better with this prompt:
Design a set of learning objectives for a class session on generative AI ethics. This is for a first-year experience class that helps students transition from high school to college. The class session will last one hour and the main goal is for students to use AI ethically in their coursework.
That’s not a highly engineered, complex prompt. I just gave Claude a little context. The results were much stronger and tailored to my situation. If you’re curious, I’ve added the actual output from each prompt at the end of this article.
The contextualized prompt produced a result that was tailored to the specific audience and goal. It was also much more practical and would be easier to implement. It’s just better in almost every way imaginable.
Your Context Toolkit: Simple Frameworks That Work
You don’t need to get carried away with providing context. Just treat AI like it was an expert who had little knowledge of your context. Here’s one framework you can use to give adequate context.
Your context should include:
WHO you are: Your role, expertise level, and any relevant background (e.g., "I'm a biology professor teaching an introductory course")
WHAT you want: Your specific goal or desired outcome (e.g., "I need to create exam questions that test critical thinking")
WHO it's for: Your audience and their characteristics (e.g., "The students are first-year undergraduates with varying levels of science background")
Use the Who-What-Who context framework as part of my Context-Goal-Task prompting framework.
CONTEXT: The situation and background information
GOAL: What you want to accomplish
TASK: The specific thing you want the AI to do
Here’s an example that puts all of this together:
I'm an economics professor teaching principles of macroeconomics to a mix of business majors and general education students. Many of my students have never experienced significant inflation before this year. I want to help students understand how economic concepts affect their personal lives. Create three real-world scenarios showing how inflation impacts different types of people.
“What” in the context framework has some overlaps with “Goal” in the CGC framework. The intent is for goal to be a bit broader, but don’t get hung up in making these frameworks overly precise. Just include what makes sense to you at the time. Frameworks are just guidelines and convenient ways to organize things, they’re not laws.
Feel free to completely ignore the frameworks if you like to do things differently. The important thing to remember is that including context will dramatically improve your results.
Put It Into Practice: Your Next AI Conversation
The next time you use ChatGPT, Claude, or any other AI tool, take a moment to provide context before diving into your request. Even just 30 seconds spent explaining who you are, what you want, and who it's for can save you multiple rounds of back-and-forth with the AI.
Start small - try adding context to your very next AI interaction and see the difference for yourself. You might be surprised at how much more targeted and useful the responses become.
Have you tried using context in your AI prompts? I'd love to hear about your experiences. Drop me a line at craig@AIGoesToCollege.com or leave a comment below. And don't forget to check out the AI Goes to College podcast, which I co-host with Dr. Robert E. Crossler, available at https://www.aigoestocollege.com/follow.
Until next time, here's to better AI conversations!
See For Yourself: The Before and After
(Note: The output formatting looks a bit unusual because Claude uses markdown language, and the # symbols indicate heading styles.)
No context prompt
# Learning Objectives: Ethics of Generative AI
## Cognitive Domain Objectives
### Knowledge and Comprehension
1. Understand Fundamental Ethical Concepts
- Define key ethical principles as they apply to generative AI technologies
- Explain the core ethical challenges posed by large language models and generative AI systems
- Describe the fundamental differences between traditional software ethics and AI ethics
### Analysis and Critical Thinking
2. Analyze Ethical Implications
- Critically evaluate the potential positive and negative societal impacts of generative AI technologies
- Identify and deconstruct complex ethical dilemmas arising from AI-generated content
- Compare and contrast different ethical frameworks for assessing AI technology
### Synthesis and Evaluation
3. Develop Ethical Decision-Making Skills
- Construct nuanced arguments about the responsible development and deployment of generative AI
- Evaluate real-world case studies involving ethical challenges in generative AI
- Propose comprehensive ethical guidelines for AI development and use
## Practical Domain Objectives
### Ethical Awareness and Reflection
4. Recognize Ethical Challenges
- Identify potential biases in generative AI systems
- Understand the implications of AI-generated content on:
* Intellectual property
* Privacy
* Misinformation
* Labor markets
* Social interactions
### Practical Application
5. Apply Ethical Reasoning
- Develop strategies for mitigating ethical risks in AI development
- Create ethical assessment frameworks for generative AI technologies
- Design responsible AI usage guidelines for professional and personal contexts
## Attitudinal Domain Objectives
6. Foster Ethical Consciousness
- Cultivate a critical and reflective approach to emerging AI technologies
- Develop a sense of social responsibility regarding AI development and deployment
- Appreciate the complex interplay between technological innovation and ethical considerations
## Skill Development Objectives
7. Develop Analytical and Communication Skills
- Articulate complex ethical considerations in accessible language
- Engage in constructive dialogue about the ethical implications of generative AI
- Critically assess and communicate the nuanced impacts of AI technologies
## Specific Competency Objectives
8. Technical and Ethical Competencies
- Understand basic technical mechanisms that give rise to ethical challenges in AI
- Recognize the limitations and potential risks of generative AI systems
- Develop preliminary skills in ethical auditing of AI technologies
## Assessment Preparation Objectives
9. Prepare for Ethical Reasoning
- Develop a framework for ethical decision-making in AI contexts
- Practice applying ethical principles to complex, real-world scenarios
- Build confidence in addressing challenging ethical questions related to generative AI
## Conclusion
By the end of this session, students will have developed a comprehensive understanding of the ethical dimensions of generative AI, combining theoretical knowledge, practical skills, and a critical perspective on technological innovation and its societal implications.
Results with context
# Generative AI Ethics: Learning Objectives for First-Year Students
## Course Context
Course Type: First-Year Experience Seminar
Session Duration: 60 minutes
Primary Goal: Develop ethical understanding and responsible use of generative AI in academic work
## Cognitive Domain Learning Objectives
### Knowledge Acquisition
1. Understand AI Fundamentals
- Define generative AI and explain its basic operational principles
- Identify the key differences between AI-generated content and human-created work
- Recognize the technological capabilities and limitations of current generative AI tools
### Comprehension and Analysis
2. Ethical Reasoning
- Critically evaluate the ethical implications of using generative AI in academic settings
- Distinguish between appropriate and inappropriate uses of AI in coursework
- Analyze potential consequences of misusing AI-generated content
### Application and Practical Skills
3. Responsible AI Usage
- Develop strategies for ethically integrating AI tools into academic research and writing
- Create a personal framework for transparent and honest AI use in coursework
- Demonstrate ability to cite and acknowledge AI-assisted work appropriately
### Reflective Learning
4. Self-Awareness and Academic Integrity
- Reflect on personal motivations for using generative AI
- Assess individual learning goals and how AI can support (not replace) academic growth
- Develop a nuanced understanding of academic integrity in the age of AI
## Skill-Based Learning Objectives
### Technical Competence
5. AI Tool Literacy
- Practice identifying AI-generated content
- Learn basic techniques for verifying the originality of academic work
- Understand university policies related to AI use in coursework
### Communication and Transparency
6. Ethical Communication
- Develop skills in transparently discussing AI use with professors
- Create clear documentation of AI tool usage in academic assignments
- Understand the importance of intellectual honesty
## Behavioral Learning Objectives
### Ethical Decision-Making
7. Responsible Academic Practices
- Cultivate a personal commitment to academic integrity
- Build confidence in independent thinking and original work
- Recognize AI as a supportive tool for learning, not a replacement for critical thinking
### Professional Development
8. Future-Ready Skills
- Understand the evolving landscape of AI in professional and academic environments
- Develop adaptable approaches to emerging AI technologies
- Prepare for responsible AI use in future academic and professional contexts
## Assessment Indicators
Students will demonstrate achievement of these objectives by:
- Participating actively in class discussions
- Completing a reflective writing exercise on AI ethics
- Developing a personal AI usage guidelines document
- Engaging in peer-to-peer discussions about ethical AI use
## Key Takeaway
By the end of this session, students should view generative AI as a powerful tool that requires thoughtful, ethical, and transparent use in their academic journey.