Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Skynet to Schoolnet

Skynet to Schoolnet

Exploring the role of AI in education and some experiments, lessons, challenges and risks.

Matthew Draycott

July 10, 2024
Tweet

More Decks by Matthew Draycott

Other Decks in Education

Transcript

  1. SESSION OUTLINE • Understanding AI • AI in Education •

    Practical Applications of AI in the Classroom • Live Demonstration of AI Tools • Challenges • Q&A
  2. AI IN EDUCATION Research & Writing Curriculum Planning & Design

    Teaching, Learning, and Assessment AI Tutoring Automated Workflows, Personalised Learning & Assisted Marking
  3. TO GET YOU STARTED: Category Platforms Examples of Use Multimodal

    ChatGPT, Copilot, Claude, Gemini Creation, summarisation, search, checking, analysis, design and ideation. Art and Graphic Design Firefly, Midjourney, DALL-E, Stable Diffusion, Canva Creative content generation, digital art, concept design. Transcription Fathom AI, Sonix, Grain Transcription services, note taking, summarisation. Writing Assistance Grammarly, ProWritingAid, Wordtune, QuillBot AI, Hemingway Grammar checking, style enhancement, readability improvements. Academic Semantic Scholar, Consensus, Scite, Elicit, Research Rabbit Search, brainstorming, curation, analysis.
  4. PRACTICAL EXAMPLES TO EXPLORE • Paper Summarising & Analysis •

    Curriculum Design • Task Design • Tutoring
  5. PROMPT ENGINEERING?? There is no real magic to writing prompts,

    but there are some general guidelines for prompting that seem to help the AI return more useful completions to you.. One such structure is: • Role (act as…) • Task (summary of what the AI needs to do) • Requirements (what the completion needs to include, contain, be, etc.) • Instructions (what the AI should do to act on the prompt)
  6. WHAT I’VE LEARNT • AI doesn’t reason, it offers possible

    answers to our prompts based on available evidence, probability and analysis. • AI hallucinates, in some contexts more than others. • AI can, and will, fixate depending on its training data. • AI will be the working partner you shape it to be. • AI doesn’t empathise.
  7. CHALLENGES • Plagiarism and cheating: AI-generated content can make it

    easier for anyone to create assignments, essays, exams, and even publications.
  8. CHALLENGES • Plagiarism and cheating: AI-generated content can make it

    easier for anyone to create assignments, essays, exams, and even publications. • Critical thinking: With AI providing quick answers, people may become overly reliant on AI-generated information, hindering their ability to develop critical thinking and problem-solving skills.
  9. CHALLENGES • Plagiarism and cheating: AI-generated content can make it

    easier for anyone to create assignments, essays, exams, and even publications. • Critical thinking: With AI providing quick answers, people may become overly reliant on AI-generated information, hindering their ability to develop critical thinking and problem-solving skills. • Teaching, learning and assessment methods: Educators need to adapt our methods to incorporate, address, and challenge AI platforms effectively.
  10. CHALLENGES • Plagiarism and cheating: AI-generated content can make it

    easier for anyone to create assignments, essays, exams, and even publications. • Critical thinking: With AI providing quick answers, people may become overly reliant on AI-generated information, hindering their ability to develop critical thinking and problem-solving skills. • Teaching, learning and assessment methods: Educators need to adapt our methods to incorporate, address, and challenge AI platforms effectively. • Bias: AI models can unintentionally perpetuate biases present in their training data. Educators need to be aware of these biases and teach students how to identify and challenge them.
  11. CHALLENGES • Plagiarism and cheating: AI-generated content can make it

    easier for anyone to create assignments, essays, exams, and even publications. • Critical thinking: With AI providing quick answers, people may become overly reliant on AI-generated information, hindering their ability to develop critical thinking and problem-solving skills. • Teaching, learning and assessment methods: Educators need to adapt our methods to incorporate, address, and challenge AI platforms effectively. • Bias: AI models can unintentionally perpetuate biases present in their training data. Educators need to be aware of these biases and teach students how to identify and challenge them. • Ethical concerns: Educators and researchers also need to consider privacy concerns, data security, copyright and the impact on student autonomy.
  12. CHALLENGES • Plagiarism and cheating: AI-generated content can make it

    easier for anyone to create assignments, essays, exams, and even publications. • Critical thinking: With AI providing quick answers, people may become overly reliant on AI-generated information, hindering their ability to develop critical thinking and problem-solving skills. • Teaching, learning and assessment methods: Educators need to adapt our methods to incorporate, address, and challenge AI platforms effectively. • Bias: AI models can unintentionally perpetuate biases present in their training data. Educators need to be aware of these biases and teach students how to identify and challenge them. • Ethical concerns: Educators and researchers also need to consider privacy concerns, data security, copyright and the impact on student autonomy. • Evaluation of AI tools: It is challenging to identify and evaluate the quality, effectiveness, and appropriateness of various AI tools available for educational use, so we need to work together and share our practice.
  13. WHERE NEXT… In education, we need to consider: • How

    to prepare learners for a world in-which interaction with AI systems will be an integral part of their working lives. • How to explore with them, the ethics of using these platforms and what responsible usage looks like. • How to design for/with AI. • How to assess learning for/with AI. • How to offer something which AI cannot…
  14. A photorealistic scene of the Terminator, resembling Arnold Schwarzenegger, teaching

    a university class. The Terminator is standing at the front of a modern lecture theatre, wearing a leather jacket and sunglasses, pointing at a chalkboard with complex diagrams and equations. At least half his face is robotic. The lecture theatre has tiered seating with rows of desks and chairs. The classroom is filled with diverse university students, attentively listening and taking notes. The background includes typical lecture theatre elements like a podium, large windows letting in daylight, and academic posters on the walls. The overall atmosphere is academic. The camera is positioned at the rear of the class, taking a photo down into the class looking toward the terminator. THANKS (AI)