<aside> <img src="/icons/help-alternate_gray.svg" alt="/icons/help-alternate_gray.svg" width="40px" /> These are slides and notes to accompany a session on AI for Reading, Writing and Note taking. Please email [email protected] for comments.

</aside>

Slides

https://docs.google.com/presentation/d/1OpBtPW89bTgEQBa1QgcU5lbnTBHBmpBk/edit?usp=sharing&ouid=115960947538376875049&rtpof=true&sd=true

Key points

AI Applications: AI can be used for reading, writing, note-taking, and project planning, but its output should be carefully assessed.

• Evolution of AI: AI has progressed from expert systems to generative AI, with large language models (LLMs) like ChatGPT at the forefront.
• LLM Variations: LLMs differ in capabilities and costs, with options for free access or paid subscriptions.
• AI Capabilities and Limitations: AI can perform tasks like translation, text generation, and code generation, but it also has limitations such as hallucination and unpredictability.
• Jagged Technological Frontier: AI models exhibit varying capabilities, leading to a "jagged technological frontier" where some tasks are easy and others remain challenging but this does not map onto the expectations of humans.
• Effective AI Use: Understanding AI's strengths and weaknesses, experimenting with prompts, and critically evaluating output are crucial for effective use.
• Key Considerations: When working with AI, consider factors like capability mismatch, hallucination, prompt sensitivity, context window dependence, and variability across tools and models.
• Experimentation and Refinement: Experiment with prompts, examples, personas, and iterations to refine AI-generated content.
• Context Window: The context window includes your prompt, model-generated text, and background instructions, impacting AI's responses.
• Tools and Resources: Explore various AI tools like ChatGPT, Claude, Gemini, and others, each with strengths and weaknesses.

AI Generated Summaries

<aside> <img src="/icons/help-alternate_gray.svg" alt="/icons/help-alternate_gray.svg" width="40px" /> These summaries were generated by Gemini and Copilot directly inside the slides.

</aside>

Summary by Gemini

• AI technology has evolved from expert systems to generative AI, with large language models (LLMs) like ChatGPT being prominent examples.
• LLMs vary in capabilities and costs, with some offering free access and others requiring a subscription.