I’ve never used Claude before, this time I’ve chosen to use Claude to try and generate an answer. The prompt I entered was “In 100 words, Is melatonin medicine a health drug?”

Claude’s response is

Melatonin is not technically a “drug” but rather a hormone supplement that’s available over-the-counter. It mimics the natural sleep hormone produced by your brain’s pineal gland. While regulated as a dietary supplement in the United States, some countries classify it as a prescription medication. Despite not being a traditional pharmaceutical drug, melatonin can have physiological effects on sleep patterns and circadian rhythms. It’s generally considered safe for short-term use, though its regulation and quality control may be less stringent than prescription medications. As with any supplement, it’s wise to consult a healthcare provider before starting melatonin, especially if you have existing health conditions.

“In 100 words, Is melatonin medicine a health drug?” Claude 3.5 Sonnet, 20 Jun, Anthropic, 7 Oct. 2024, claude.ai/chat.

Claude quickly generated a well-organized answer for me. The AI answers were mostly accurate and there is not much room for AI to make things up when answer questions that have been around for many years. But I’m sure that if I tried to go deeper and ask for certain ideas. I believe that if I try to ask about certain points in depth, the AI will try to follow my point of view even if I am wrong.

Copilot’s SRMA analysis

Substitution
At this level, DALL-E 2 serves as a direct substitute for traditional tools with no functional change. For example:

Illustrations: Using DALL-E 2 to create images for educational materials instead of manually drawing or sourcing images.
Visual Aids: Generating images to accompany text in presentations or worksheets, replacing clipart or stock photos.

Augmentation
Here, DALL-E 2 provides functional improvements over traditional methods. Examples include:

Enhanced Visuals: Creating high-quality, customized images that are more relevant and engaging for students, improving comprehension and retention.
Interactive Learning: Allowing students to input their own descriptions and see the generated images, making learning more interactive and personalized.

Modification
At this stage, DALL-E 2 allows for significant task redesign. Examples include:

Creative Projects: Students can use DALL-E 2 to generate images for creative projects, such as storybooks or presentations, fostering creativity and critical thinking.
Collaborative Learning: Groups of students can collaborate on projects by generating and discussing images, enhancing teamwork and communication skills.

Redefinition
This level involves creating new tasks that were previously inconceivable. Examples include:

Virtual Field Trips: Using DALL-E 2 to generate images of historical events, distant places, or abstract concepts, providing immersive learning experiences that were not possible before.
Exploratory Learning: Encouraging students to explore and visualize complex concepts or scenarios by generating images based on their own inquiries and hypotheses.
DALL-E 2 can significantly enhance the learning experience by making it more engaging, interactive, and creative

“Create a SAMR analysis about the use of DALL-E 2 for learning.” Copilot, 1 Nov. 2023, Microsoft, 8 Oct. 2024, copilot.microsoft.com.

The analysis of DALL-E 2 for learning generated by Copilot shows how well DALL-E 2 facilitates formal learning. The model shows me that the DALL-E 2 tool generates creative images from text to present ideas or attempts to present abstract or conceptual ideas. This greatly enriches the richness of the images acquired during the learning process. It is certain that DALLE-2 is a very effective learning tool. However, there are omissions in the Copilot analysis, such as the failure to write about potential problems with DALL-E 2, such as copyright issues, and the dangers of relying on such a generative AI tool.

Reflections

  • 1.Try out creating an H5P interactivity using ChatGPT – what did it help with and what needs adjusting[1]?

I use Chatgpt to generate questions, the formatting is fine, which makes the formatting very easy, but the answers are not guaranteed, for example, this, a simple arithmetic question, the correct answer should be 197.8 rounded up to 198. But it turns out that it generates an answer of 196. So even though the formatting is correct, the content checking is a necessary step. They are processing language rather than understanding text[1].

  • 2. How accurate or successful were the learning objects you created using the AI tools[1]?

The success rate of creating learning objects with chatgpt depends on the direction you choose, if it’s something like a math problem, the success rate isn’t very high, I think it might be version related, because I’ve been using the free version of gpt3.5. But if it’s something like asking for definitions, the success rate is much higher. So I think AI tools are very effective for learning definitions and theories, while practice questions should be used with caution.

  • 3.What ethical concerns do you have (or not have) about the use of some of these tools[1]?

I basically just use AI tools for learning not for creating or showing to other people. So I don’t have any concerns about copyright, and I think that’s more of a concern for AI companies. My biggest concern is the disinformation. The short-form video space is hot right now, and there’s a lot of short-form video that can be generated by AI, and there’s a lot of low-quality content that’s being put out on social media without being vetted by the publishers, and there’s a lot of misinformation that’s being put out there. This is my biggest concern, as it squeezes the space for normal video creators to survive on the one hand, and lacks regulation on the other.

References:

[1] Watt, Mary. “Module 2: Generative AI and Evaluating Content and Tools for Learning (Sept 29 – Oct 12).” EDCI 337, University of Victoria, 4 Dec. 2023, edtechuvic.ca/edci337/2023/12/04/module-5-generative-ai-and-evaluating-content-and-tools-for-learning-mar-17-30/. Accessed 7 Oct. 2024.

“In 100 words, Is melatonin medicine a health drug?” Claude 3.5 Sonnet, 20 Jun. 2024. version, Anthropic, 7 Oct. 2024, claude.ai/chat.

“Create a SAMR analysis about the use of DALL-E 2 for learning.” Copilot, 1 Nov. 2023, Microsoft, 7 Oct. 2024, copilot.microsoft.com.

Comment

Module 2 | bifei (opened.ca)

Image Alt Text: