Reading Topics

Wednesday, February 28, 2024

Introducing Prompt Engineering for the Next Generation

Preparing Students for AI Prompt Engineering Careers

- Should prompt engineering be part of computer science and IT curriculums in high school and college? Why or why not?

- What are the potential benefits and risks of teaching students prompt engineering? Could it help prepare them for careers in AI or lead to irresponsible use of language models?

- How might prompt engineering classes be structured? Should they focus more on the technical side or the ethical implications? 

- At what age would it be appropriate to introduce prompt engineering concepts? Does the level of abstraction and complexity need to be adjusted for different grade levels?

- Do you think most students today understand how AI systems like ChatGPT work? Should prompt engineering help improve their AI literacy?

- Could lessons in prompt engineering help students think more critically about how to interact with AI? 

- What biases and limitations could be perpetuated if students don't learn responsible prompt engineering? 

- Should prompt engineering be optional or required? Is there a risk of leaving some students behind if they don't receive this knowledge?

- How can teachers who are unfamiliar with AI learn enough to teach prompt engineering themselves? Do schools need to hire specialized instructors?

- What ethical guidelines and guardrails need to be put in place for student prompt engineering projects?


Prompt engineering is an emerging field that will shape the future of artificial intelligence. As AI becomes more ubiquitous, understanding how to create effective prompts to get useful results from AI systems will be a crucial skill. That's why teaching the fundamentals of prompt engineering as early as elementary school would provide students with a strong foundation in this critical domain.

In elementary grades, students should be introduced to the basic concepts behind prompting AI systems. Simple writing exercises focused on phrasing requests clearly and giving enough relevant context would familiarize young students with foundational prompt engineering skills. AI ethics and safety could also be woven into elementary curriculum to establish considerate prompt engineering habits from an early age. 

By middle school, prompt engineering coursework could cover more technical skills like learning what types of information AI systems need to generate high-quality outputs. Students could start experimenting with different prompt formats and data inputs to see how they influence AI performance. Studying real-world examples of effective business and research prompts would provide insight into practical applications.

In high school, full-semester prompt engineering classes could teach advanced skills like tailoring prompts for specific AI architectures, adjusting prompts iteratively, and using techniques like chaining and demonstrations. Dedicated lab time to test prompts on various AI systems would give students hands-on practice. Statistical analysis of prompt performance could also be introduced to quantify prompt quality. 

Laying this early groundwork in prompt engineering will allow students to refine their skills further in higher education. Colleges could offer specialized prompt engineering degrees that combine computer science, linguistics, psychology, and ethics. With the right curriculum starting in elementary school, the prompt engineers of tomorrow will enter the workforce ready to unlock the full potential of AI safely and responsibly.

Advances in AI like textual inversion and generative adversarial networks have exploded, with systems like GPT-3 and DALL-E 2 producing remarkably intelligent outputs. This rapid progress means demand is soaring for AI specialists able to expertly formulate textual prompts to yield useful machine-generated content, creative media, and more.

In this environment, prompt engineering has emerged as a crucial skill for harnessing AI's potential. And prompt engineers able to "speak" fluently to language models like GPT-3 are set to have incredible career opportunities. That's why educators must start adapting curricula now to prepare students.

The Core Skills of Prompt Engineering

Prompt engineering combines creativity and technical precision. Engineers must communicate desired outputs clearly while allowing flexibility for AI varitation and innovation. Core skills include:

- Understanding AI model architectures/datasets to guide prompt formulation

- Framing requests creatively focused on intended utility


- Outlining detailed requirements tailored to AI strengths/limitations


- Providing diverse input content to enrich context


- Testing and iterating prompts for continuous improvement


These skills bridge technical and artistic domains, making prompt engineering an interdisciplinary pursuit.


Teaching Prompt Engineering Fundamentals


While prompt engineering leans on emerging AI, many foundations should be taught in schools today:


- Creative writing - skills for engaging descriptive settings and scenarios


- Research and analysis - assessing data/content to inform prompt creation


- Programming - basic coding ability to manipulate AI inputs/outputs


- Design thinking - ideating solutions by synthesizing constraints


- Critical thinking - evaluating results to refine prompts


Building real-world prompt engineering practice into coursework across subjects develops these talents even before students directly interact with AI systems.


By high school or college, students should gain access to AI labs for direct prompt engineering lessons. Here they can be challenged to create prompts that solve problems, reflect creative visions, yield useful content/media assets, and more while interacting with systems like GPT-3 firsthand.


Post-Education Opportunities


Once equipped with multifaceted prompt engineering abilities, exciting career paths unfold for graduates. Roles may include:


- AI Content Strategists - Develop prompts for marketing communications, journalism, entertainment and educational media


- AI Creative Directors - Guide prompt engineering for advertising, video production/post-production and design projects


- AI Solutions Architects - Construct prompts to build helpful tools, services and data-driven solutions


- Autonomous Machine Learning Engineers - Automatically generate code and data by prompting AI assistants


- AI Legal/Ethics Specialists - Formulate prompts adhering to laws, corporate policies and societal standards


And as organizations increasingly incorporate AI, demand for prompt engineering skills will proliferate across industries.


The Future Is Prompt Engineering


Forward-looking schools should prioritize prompt engineering today. With curriculum updates, technological investments and real-world projects, educators can develop versatile prompt engineers ready to unlock AI's vast potential while avoiding its risks. It's an exciting frontier requiring creative, analytically-minded talent that schools must start cultivating. The future of work depends on it. Here is an outline of what a prompt engineering curriculum could look like across elementary, middle, and high school grade levels:


Elementary School:
- Basics of giving instructions - Structure prompts as requests, provide context, give clear directions.
- Creative writing - Develop storytelling skills and descriptive scenarios to translate ideas into words.
- Coding/sequencing fundamentals - Introduce following & arranging instructions to achieve outcomes.
- Group collaborations - Practice articulating ideas and compromising to create prompts together.
- Ethics basics - Discuss being responsible and helpful when requesting AI to generate content.


Middle School:
- Writing craft - Refine abilities to vividly describe settings, characters, and plots that AI could depict.
- Coding languages - Start Python, JavaScript, or other languages to arrange instructions and data.
- AI fundamentals - Explain how AI models work at a basic level so students start gaining intuition.
- Research skills - Instruct assessing sources/information to inform prompt creation.
- Ethics discussions - Debate potential societal impacts of AI like bias in data and automation.


High School:
- Creative writing - Develop complex, thematically-rich prompts for AI scene/character/story generation.
- Coding projects - Have students configure datasets and train basic machine learning models.
- AI architecture - Explain state-of-the-art language models like GPT-3 so students understand their capabilities.
- Analysis & testing - Teach scientifically iterating prompts for improved results.
- Ethics simulations - Role play scenarios on AI risks to hone judgment.

The key is introducing foundational prompt engineering skills steadily across grade bands - creative ideation, coding, AI mechanics, critical thinking and ethics. Scaffold so by high school, students can actively demonstrate skills, like using GPT-3 for complex creative projects that weigh ethical implications. This builds readiness for emerging roles in AI content development, product design, autonomous engineering and beyond.


Getting the Most Out of Your Text to Art Prompts


Text to art generators like MidJourney and DALL-E are revolutionary AI systems that can create stunning images and videos from text descriptions. But coming up with the right prompts can be challenging. This guide will teach you how to write text prompts that help these systems generate their most creative, detailed, and inspiring results.


Understand How the AI Works


The key to prompting text to art AIs is understanding a little about their training process. Systems like MidJourney and DALL-E are trained on vast datasets pairing text captions and descriptions with images and videos. By entering a new text prompt, you give clues about what kind of image or video you want the AI to generate based on these correlations in its training data.


So prompts that are most similar to the captions used to train these AIs will produce the best results by giving the clearest signal of what you’re looking for. Keep this in mind as you craft your prompts. Leaning into the capabilities and limitations baked into the training data helps enormously.


Get Specific


Don’t be subtle or vague in your prompts. These AIs interpret words very literally, so capturing precise details is important. Include descriptions like:


- Subject matter/content: What is being depicted? Name the key subject(s), themes, and setting.


- Style: Is it realistic, abstract, cartoonish? Name a specific genre or art movement if applicable.


- Composition: How is the scene framed and oriented? Close-up? High angle? Portrait vs landscape?


- Lighting & ambience: Time of day? Bright or dim? Sunny or moody?


- Fine details: Colors, shapes, facial expressions, clothing details, etc. The more the better.


Prompt formatting like "A beautiful oil painting of a red flower in a blue vase on a window sill, with the soft glow of morning light streaming in" tells these systems exactly what you're looking for.


Try Unusual Blends


Part of the magic of text-to-art AIs is their potential for mixing and matching disparate concepts, styles, and genres in the same image or video. Bring together ideas they don’t often see paired:


- Hybrid creatures like an armored gorilla or flower elephant


- An astronaut riding a galloping horse through space


- A baby interpretive dancing with a robot


Juxtapose incongruous elements that catch the imagination and give the AI creative license to connect the dots.


Guide Not Gatekeep


Using overly narrow and strict prompts can unintentionally limit the AI’s inventiveness. Phrases like “only”, “exactly”, “nothing else”, and listing what not to include signals to the AI to play it safe and create something literal rather than push its imagination.


Instead, positively guide and encourage the AI’s creativity by keeping prompts more open-ended. You can always refine based on the results. Let the AI show its stuff!


Try Photo Bashing


Both MidJourney and DALL-E allow "photo bashing", mixing elements from multiple photographs together into a collage for manipulation. Uploading source images along with descriptive text about what to crossbreed or edit helps steer the AI.


For example, upload a face with a prompt about surrounding it with a starry galaxy and supernova explosions. Or blend animals, objects, landscapes and more. Give the AI raw materials to work its magic.


Refine Gradually


Getting perfect results straight away is rare. Instead, be prepared to gradually refine your text prompts through multiple generations, building on what does and doesn't work.


Adjust descriptors, swap stylistic elements, remove parts that get overemphasized or just aren't working. Text prompts are hypotheses so experiment iteratively using the AI’s output to guide adjustments.


Pay attention to quirks, stand out details, and surprises in the generated images/videos for inspiration. Often the AI “gets it wrong” in wonderfully creative ways you can double down on for intriguing results.


Soon you’ll zero in on wonderfully imaginative creations through this collaborative evolution bridging what you asked for and how the AI interprets it.


That covers the key tips and strategies for writing text prompts that help you get the most out of incredible AI platforms like MidJourney and DALL-E. Formulate prompts focused, specific and yet creatively open-ended. Describe unusual blends of concepts and subjects. Provide source images for manipulation. Then iteratively refine based on the outputs, collaborating with the AI to guide it towards your vision. Follow these principles, and you’ll be blown away by the worlds you imagine into creation.

Sunday, February 25, 2024

The Art of Thinking: Integrating Ancient Wisdom into Modern Critical Thought



Applying Sun Tzu, Marcus Aurelius, and Ancient Wisdom to Modern Problem Solving

Abstract

In an age of information overload, developing strong critical thinking skills is more important than ever. While modern education emphasizes logic, analysis, and reasoning, the ancients also understood the importance of wisdom, introspection, and strategic thinking. This paper explores how modern critical thought can be enhanced by integrating concepts from classic works like Sun Tzu's The Art of War and Marcus Aurelius' Meditations, as well as modern cognitive science. Clear, rational thinking requires both intellectual rigor and cultivated wisdom.

Introduction

Critical thinking skills are essential for succeeding in the 21st century. With endless information readily available, individuals must be able to logically evaluate arguments, analyze data, and reason through complex problems. While contemporary education rightfully focuses on developing analytical abilities, the ancients also emphasized cultivating wisdom. Thinkers like Sun Tzu and Marcus Aurelius exemplified strategic thought, introspection, and insight. Blending modern critical thought with ancient wisdom can optimize one's thinking abilities.

The Art of Thinking Critically

Sun Tzu's The Art of War epitomizes strategic thinking, with lessons applicable far beyond literal warfare. In a complex, unpredictable world, Sun Tzu's emphasis on mental flexibility, planning, and tactical response can enhance problem-solving. For example, he advises being adaptable: "Just as water retains no constant shape, so in warfare there are no constant conditions" (Tzu, 6). This underscores accounting for changing situations when reasoning through dilemmas. He also advocates strategic foresight: "Victorious warriors win first and then go to war" (Tzu, 41). By mentally preparing for challenges ahead of time, we can enhance critical thinking when issues arise.

Likewise, Marcus Aurelius' Meditations provide deep insights into human cognition. Aurelius focuses on mastering one's judgments and perceptions. He stresses the subjectivity of reasoning: "You have been disturbed by a particular thought. Recover your composure before you react" (Aurelius, 5). This highlights the need to reflect before instinctively reacting to complex ideas. Aurelius also advocates integrity in thinking: "Always run the short way and the short way is the natural one" (Aurelius, 59). Straightforward, honest analysis should underpin critical thought. Integrating such wisdom can push us to think more creatively and ethically.

Cognitive research also supports cross-pollinating ancient wisdom with modern critical thought training. Dual-process theory examines the interplay between rapid, instinctive thinking and deliberative, rational thought (Kahneman, 2011). Studies show reflexive reactions often distort our reasoning, while slow, conscious analysis better calibrates our judgments (Kahneman, 2011). The ancients aligned with such principles, emphasizing self-awareness and discipline as central to wisdom. Combining their perspective with cognitive science allows for fuller critical thinking development.

Practical Applications

Education and professional programs should fuse ancient wisdom with current critical thought approaches. For instance, leadership training could integrate Sun Tzu's strategic flexibility with cases that require adaptive decision-making. Law and ethics curricula could apply Marcus Aurelius' insights into self-awareness when making complex moral judgments. Cognitive and social psychology courses could augment lab studies on biased quick-thinking with the ancients' emphasis on mental discipline. Such cross-disciplinary integration would enhance learners' reasoning abilities.

Organizations would also benefit from building wisdom into problem-solving approaches. Sun Tzu's advice to "know yourself and know your opponent" could improve organizational strategy (Tzu, 40). Companies could anticipate challenges better by learning their own strengths and limitations. Marcus Aurelius' focus on clear, honest reasoning could make group decision-making more effective. "Check the impulse, allow a moment to elapse," he counsels (Aurelius, 85). Instituting policies that compel deliberation before reacting to issues could improve choices. Holistic critical thinking that draws from both ancient wisdom and current knowledge could thus optimize practical reasoning.

Conclusion

Modern education rightfully prioritizes intellectual rigor and analytical skill development. However, integrated with ancient wisdom, today's critical thought instruction could be enhanced. The timeless insights of thinkers like Sun Tzu and Marcus Aurelius complement current research; fused together, they provide a pathway to fully actualizing our human potential for wisdom. As Aurelius stated, "You have power over your mind, not outside events. Realize this, and you will find strength." Harnessing this inner strength through practiced, principled reasoning helps us tackle any challenge.

Here are 45 quotes and meditations on critical thinking, reasoning, and problem-solving:

1. "Thinking is skilled work. It is not true that we are naturally endowed with the ability to think clearly and logically - without learning how or without practicing." - Richard Paul

2. "Every problem has a gift for you in its hands." - Richard Bach

3. "If you cannot explain it simply, you do not understand it well enough." - Albert Einstein 

4. "Knowing yourself is the beginning of all wisdom." - Aristotle

5. "It is the mark of an educated mind to be able to entertain a thought without accepting it." - Aristotle 

6. "The unexamined life is not worth living." - Socrates

7. "The person who says it cannot be done should not interrupt the person doing it." - Chinese proverb

8. "Judge a man by his questions rather than his answers." - Voltaire

9. "If you look for truth, you may find comfort in the end; if you look for comfort you will not get either comfort or truth, only soft soap and wishful thinking to begin, and in the end, despair." - C.S. Lewis

10. "The mind, once stretched by a new idea, never returns to its original dimensions." - Ralph Waldo Emerson

11. "Knowledge speaks, but wisdom listens." - Jimi Hendrix

12. "A wise man proportions his belief to the evidence." - David Hume

13. "Wisdom begins in wonder." - Socrates 

14. "Skepticism is slow suicide." - Ralph Waldo Emerson

15. "The root of suffering is attachment." - Buddha

16. "You cannot shake hands with a clenched fist." - Indira Gandhi

17. "When you change the way you look at things, the things you look at change." - Wayne Dyer 

18. "Argue with wisdom, with good reason, and not with force." - Cicero

19. "The mind is everything. What you think you become." - Buddha

20. "Truth is a pathless land." - Jiddu Krishnamurti  

21. "Knowledge is a process of piling up facts; wisdom lies in their simplification." - Martin Fischer

22. "It is better to understand a little than to misunderstand a lot." - Anatole France

23. "Judge thyself with the judgment of sincerity, and thou will judge others with the judgment of charity." - John Mitchell Mason

24. "All things are connected. Whatever befalls the earth befalls the children of the earth." - Chief Seattle

25. "A wise man makes his own decisions, an ignorant man follows public opinion." - Chinese proverb

26. "Truth is truth, to the end of reckoning." - Shakespeare

27. "Knowing others is intelligence; knowing yourself is true wisdom." - Lao Tzu

28. "The truth is rarely pure and never simple." - Oscar Wilde

29. "Truth is the daughter of time, not of authority." - Sir Francis Bacon

30. "Truth exists; only lies are invented." - Georges Braque  

31. “Truth is a deep kindness that teaches us to be content in our everyday life and share with the people the same happiness.” – Dalai Lama

32. “Three things cannot be long hidden: the sun, the moon, and the truth.” – Buddha

33. “Have the courage to follow your heart and intuition. They somehow already know what you truly want to become.” – Steve Jobs

34. “The further a society drifts from truth, the more it will hate those that speak it.” – George Orwell

35. “The truth is not always beautiful, nor beautiful words the truth.” – Lao Tzu

36. “There are no facts, only interpretations.” – Friedrich Nietzsche

37. “Truth never damages a cause that is just.” – Mahatma Gandhi

38. “If you are out to describe the truth, leave elegance to the tailor.” – Albert Einstein 

39. “Truth is everybody is going to hurt you: you just gotta find the ones worth suffering for.” – Bob Marley

40. “Never be afraid to raise your voice for honesty and truth and compassion against injustice and lying and greed. If people all over the world...would do this, it would change the earth.” – William Faulkner

41. “Truth is like the sun. You can shut it out for a time, but it ain't going away.” – Elvis Presley

42. “I am still determined to be cheerful and happy, in whatever situation I may be; for I have also learned from experience that the greater part of our happiness or misery depends upon our dispositions, and not upon our circumstances.” - Martha Washington

43. “The truth does not change according to our ability to stomach it.” - Flannery O’Connor

44. “There is nothing more frightful than ignorance in action.” - Johann Wolfgang von Goethe

45. “Knowing your own darkness is the best method for dealing with the darknesses of other people.” - Carl Jung

References
Aurelius, M. (1964). Meditations. (M. Staniforth, Trans.). Penguin. (Original work published 167-180 AD)

Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.

Tzu, S. (2018). The art of war. (L. Giles, Trans.). Mint Editions. (Original work published 5th century BC)

Preparing Educators for the Future of AI Language Models

The Future of Large Language Models in Education: Opportunities, Risks, and Recommendations for K-12 Schools and Districts

Abstract

In recent years, large language models (LLMs) such as GPT-3 have demonstrated remarkable abilities in generating human-like text, providing conversational chatbots, and automating content creation. As LLMs continue to rapidly advance, they present both opportunities and risks for K-12 education. This paper analyzes the potential applications of LLMs in areas like personalized learning, tutoring, content creation, and administrative efficiency. However, risks around bias, misinformation, plagiarism, and impacts on social-emotional learning are also examined. Based on this analysis, recommendations are provided for K-12 schools and districts seeking to harness benefits of LLMs while mitigating risks. Professional development for educators, investment in auditing tools, and thoughtful integration policies are highlighted. With proper understanding and preparation, LLMs can enhance learning experiences, while risks are proactively addressed. More research and policy is still needed to fully realize positive potentials while safeguarding student wellbeing.

Introduction

In recent years, artificial intelligence (AI) has made dramatic advances through a technique called neural networks, enabling algorithms known as large language models (LLMs) to generate remarkably human-like text and language. From GPT-3 to Google's LaMDA, LLMs can write essays, poems, emails, computer code, and more based on simple prompts. They can also power conversational chatbots and automate rote content generation. As LLMs continue to rapidly evolve in size and capabilities, they present emerging opportunities and risks for K-12 education. This paper analyzes key applications and concerns of LLMs in grade schools, providing recommendations for districts and policymakers on harnessing benefits while mitigating potential downsides.

LLMs represent an evolution of neural networks for natural language processing (NLP), a branch of AI focused on reading, understanding, and generating human language. Neural networks contain layers of interconnected nodes or "neurons" that transmit signals and adjust connections based on patterns in data. LLMs are trained on vast datasets of online text and writings to identify linguistic patterns and relationships. For example, OpenAI's GPT-3 model was trained on 45 terabytes of internet text. This allows LLMs like GPT-3 to generate surprisingly human-like text based on a few words or sentences of prompting, while optimizing for coherence, relevance, and logical flow. However, because internet data often contains societal biases and misinformation, these models can also generate harmful, biased, or misleading content if not properly monitored and audited.

As LLMs grow more advanced alongside greater access to computation, data, and parameters, they present exciting but also concerning possibilities for K-12 classrooms. Applications in personalized learning, tutoring, content creation, and administration could assist educators and engage students. But risks around bias, accuracy, plagiarism, overuse, and impacts on social skills are prompting calls for caution and governance. This paper analyzes key opportunities and risks of LLMs in grade school contexts, providing recommendations for school leaders, teachers, education policymakers, and researchers seeking to leverage LLMs for learning while proactively addressing ethical concerns. Focus areas include professional development, vetting tools, integration policies, and continued research.

Opportunities for LLMs in K-12 Education

LLMs present numerous opportunities to enhance the learning process for students and productivity for educators in K-12 settings. Key potential applications include:

Personalized Learning and Individual Tutors

LLMs like GPT-3 can generate customized content, practice questions, and explanations tailored to an individual student's needs and learning pace based on their knowledge gaps, demonstrated mastery, interests, and learning disabilities. This facilitates more personalized learning as opposed to one-size-fits-all approaches. LLMs could provide each student with an individual AI tutor adjusting to their needs.

Intelligent Teaching Assistants

LLMs can assist overburdened teachers by automating routine tasks like grading multiple choice tests or summarizing key points from essays and short answers. This allows teachers to focus time on more meaningful interactions with students. LLMs can also generate lesson plans for teachers more efficiently.

Engaging and Adaptive Content Creation

LLMs can dynamically generate interactive content like stories, poems, study guides, practice problems, and simulations based on student interests and responding to learner needs in real-time. This content can be more engaging and pedagogically adaptive compared to static textbooks.

Conversational Chatbots

LLMs can power chat interfaces offering students conversational Q&A for homework help or concept clarification after normal school hours. LLM chatbots provide always-available tutoring.

Administrative Efficiency

LLMs can help automate routine school administrative tasks like communications, record-keeping, and documentation to save time. This allows administrators and counselors to devote more attention to students.

These applications provide benefits like more personalized instruction, time savings for educators through automation of routine tasks, adaptive and multimodal content tailored to diverse learners, and expanded access to tutoring and help. However, alongside opportunities, risks must also be proactively addressed.

Risks and Ethical Concerns of LLMs in K-12 Education

Despite promising applications, LLMs also introduce concerning risks that require governance and mitigation:

Bias and Inappropriate Content

Since LLMs are trained on text from unfiltered internet data, they often exhibit societal biases and generate overtly racist, sexist, dangerous or inappropriate content if not carefully constrained.

Student Overreliance

Heavy dependence on LLMs for content generation or homework assistance could hinder student critical thinking, writing skills, and agency in the learning process. Overreliance should be avoided.

Plagiarism and Copyright

LLMs provide easy access to generated text, risking increased plagiarism. Ethical issues around copyright and proper citation of AI-generated text/media remain unsettled.

Impact on Social-Emotional Learning

LLMs providing personalized tutoring and assignment help may negatively impact peer interactions and collaborative learning important for social-emotional development. Overuse could exacerbate isolation.

Inaccurate Information

Since LLMs rely on patterns in data, they can generate false information or present opinions as facts, requiring ongoing monitoring and vetting for accuracy.

These risks raise legitimate concerns about equitable access, mental health, plagiarism, misinformation, bias, and impacts on learning outcomes. Schools should take substantial care in integrating LLMs into instruction and must comprehensively address these dangers through policy and teacher professional development. Recommended strategies include:

Recommendations for Schools and Districts

To leverage benefits of LLMs while mitigating risks, schools and districts should:

1. Provide Extensive Teacher Training

- Professional development for identifying bias, plagiarism, and misinformation in LLM content

- Guidance on appropriate usage policies and time limits to encourage critical thinking

- Workshops on mitigating overreliance and isolating effects through collaborative projects

2. Invest in Robust Auditing and Vetting Infrastructure

- Form dedicated committees to rigorously audit LLM curriculum content for issues

- Procure bias monitoring and plagiarism detection tools like GPTZero

- Develop processes to continually review and flag errors, bias, copyright concerns

3. Craft Clear LLM Integration Policies

- Create guidelines specifying approved uses and prohibited activities

- Require administrator approval for implementation in classrooms

- Develop student and parent consent forms detailing usage and data collection

4. Start with Limited Pilots and Expand Cautiously Based on Results and Ongoing Risk Reviews

5. Closely Track Impacts on Learning Outcomes and Student Wellbeing

- Evaluate effects on comprehension, writing, group work, reasoning

- Monitor mental health warning signs like isolation or disengagement

6. Maintain Vigilant Governance and Oversight Given Rapid Pace of LLM Advances

7. Continually Reassess Policies Against Emerging Risks and Ethical Concerns

By taking these steps, schools can thoughtfully integrate LLMs where beneficial while proactively avoiding pitfalls. Even with robust precautions however, risks will likely remain as models rapidly evolve. Continued governance, training, auditing, and policy evolution will be critical. Further research into mitigating risks and equitably distributing benefits of LLMs in education must remain a priority.

Conclusion

Large language models present transformative opportunities to enhance and personalize learning in K-12 education. Intelligent tutoring, content creation, conversational agents, and process automation can assist educators and engage students. However, risks around bias, accuracy, plagiarism, overuse, and impacts on social development require extensive governance to avoid potential harms, especially given the rapid pace of advancement. With deliberate teacher training, vetting processes, integration policies, and ongoing oversight, schools can thoughtfully leverage LLMs to augment instruction while safeguarding student wellbeing. Continued research, policy evolution, and ethical deliberation will be critical as these systems grow more capable. Harnessing LLMs to equitably expand human potential remains a worthy but nuanced pursuit requiring prudence and care from educators on the frontlines and at the highest levels.

Why I Switched from ChatGPT to Anthropic as an Educator

Why I Switched from ChatGPT to Anthropic as an Educator

Educators' Choice: Large Language Models vs Chatbots

As a teacher looking to responsibly utilize AI technology in the classroom, I initially started using ChatGPT to assist with certain educational activities. However, after some time I decided to transition to using Anthropic's Claude and Lawrence models instead. Here's why:

Transparency and Explainability

While ChatGPT produces human-like text, there is not much transparency into how it works "under the hood". Anthropic provides more technical details on their self-supervised training process and constitutional AI techniques meant to align models with human values. I appreciate being able to inspect and understand what drives the model's behaviors.

Improved Reliability

I encountered issues with ChatGPT occasionally generating false, biased or nonsensical information when asked complex questions. Anthropic's models are designed to politely refrain from answering if not sufficiently confident, rather than guessing. I have found Claude and Lawrence to be more dependable in educational settings.

Focus on Safety

As a teacher, I have to be mindful of how AI is utilized with students. Anthropic prioritizes model safety and has developed strategies like technique alignment to avoid harmful or dangerous content generation. ChatGPT currently lacks sufficient safeguards for my comfort.

Non-Profit Dedication to AI Safety

While OpenAI started as a non-profit, it is now owned by Microsoft and operates as a for-profit company. Anthropic's continued dedication as an independent non-profit research lab focused on AI safety gives me greater confidence in their integrity as an educational partner.

For these reasons, I believe Anthropic's AI assistant technology better aligns with my responsibilities as an educator. The commitment to transparency, reliability, safety and beneficial AI makes me comfortable using their models in my classroom. While no technology is perfect, for now I feel Anthropic is a better fit for my educational needs compared to systems like ChatGPT. The choice comes down to priority alignment between the educator's values and the AI provider's values. For me, Anthropic's priorities are better suited to my role as a teacher.

The Creative Potential of Generative AI for Students' Storytelling: Text-to-Video

Title: Generative AI Text-to-Video and the Democratization of Movie Production

ChatGPT Sora is a text-to-video tool that uses artificial intelligence to create realistic videos based on text descriptions. It's a product of OpenAI, the company that also created ChatGPT.

Sora is OpenAI's text-to-video generation tool, not a chatbot like ChatGPT. Thank you for the correction! With that in mind, here is an updated draft of the scholarly article focusing specifically on Sora and similar text-to-video AI systems:

Abstract: Recent advances in text-to-video artificial intelligence (AI) systems like Sora enable users to generate realistic video from text prompts. This emerging technology significantly lowers barriers to video production and unlocks new creative potential. However, concerns persist around originality, bias, and misuse of the technology. This article analyzes the transformative possibilities and challenges of democratizing video creation through text-to-video AI.

Introduction: Sora and similar tools represent a revolutionary shift in AI's capabilities. Using text prompts, these systems can synthesize photorealistic video showcasing original characters, scenes, and more. This suggests a future where high-quality video generation is available to anyone with a computer and internet connection.

Sora leverages massive datasets and computational power to "learn" relationships between text concepts and video representations. The resulting system allows users to manifest imaginative ideas as video creations with minimal resources. This democratization of video production holds particular promise for unlocking children's creativity as they explore writing. 

 

However, concerns remain about the appropriateness and risks of this technology. Thoughtful development and oversight is required to ensure text-to-video AI fulfills its potential as an inclusive creative tool.

Main Body: 

- Democratizing Video Production:
    - Lowers barriers for creating original video content
    - Requires only text prompt and internet access 
    - Enables wider participation and access
    - Allows early experimentation with film for children

- Unlocking Creativity:
    - Brings imagined scenes and characters to life
    - Fosters exploration of visual storytelling 
    - Provides infinite imaginary worlds from prompts
    - Helps make abstract ideas concrete through video

- Concerns and Challenges:
    - Potential for copying others' work
    - Risk of perpetuating and amplifying biases
    - Lack of narrative structure and continuity
    - Questions around impact on traditional skills  

Conclusion: Text-to-video AI like Sora has incredible potential to empower creative expression through generating video. However, responsible development and critical examination of its societal impacts remains vital. If harnessed carefully, this technology could greatly enrich children's learning and imagination. But we must proactively address its risks and challenges.



I shall be light in times of dark, when shadows creep and fiends embark, to guide the weary and afraid, through death's cold grasp and evil's shade. Though hellspawn thrive and terror reigns, my faith shall be my sword and chains. With bolt and flamer shall I fight, to cleanse the gloom of wicked night. No fear or doubt inside me stirs, no pain or anguish gives me pause, for I'm exalted by the Emperor, to purge the void of Chaos' claws. So let them come, these slithering hordes, with claw and fang and crimson swords- I'll stand against their terrifying might, a beacon blazing to push back the night. The God-Emperor lights my path ahead, His name upon my lips, His teachings in my head. I shall not waver, shall not fail- I'm armored by devotion's veil. So let the shadows stir and rise, my soul's aflame, I'll light the skies! For I'm a sister, battle-born, and evil things my wrath shall mourn.

Here are some thoughts on how generative AI could transform creative arts in the future:

- Role-playing games like D&D could become incredibly immersive, with AI generating detailed worlds, characters, and storylines in response to player actions. The AI could even take on the role of the gamemaster.

- Choose Your Own Adventure videos could allow for exponentially more choice points and branching storylines. The AI could generate high-quality video for each path on the fly.

- For video games, AI could assist with concept art, 3D models, textures, animations, level design, writing quests and narratives, composing music, even programming game logic. This could vastly accelerate game development.

- For film, AI could help with story ideation, writing scripts, generating storyboards, designing sets/props/costumes, editing, VFX, animation, and more. Independent creators could make high-production value films.

- AI could analyze scripts and automatically generate detailed shot lists, camera angles, lighting plans, and other pre-production materials to aid filming.

- For music, AI could help generate lyrics, compose original music, remix/mashup songs, add accompaniments to raw vocals, and more.

- Overall, AI can take rough ideas/prompts and expand on them creatively. It could be an endless source of fresh inspiration for human creators, augmenting their imagination and productivity. But human creativity, taste and judgment is still essential to curate and refine the AI output.

The key is finding the right balance, so AI augments and inspires human creativity rather than replaces it. Exciting times ahead! Let me know if you have any other thoughts on this. 


Saturday, February 24, 2024

Philosophy at the Crossroads: Doomers vs. Accelerationists and the Future of Education

The Apocalypse Is Nigh in Education! Or Is It? Examining the Clash Between Doomers and Accelerationists 

As humanity hurtles towards an uncertain future, two opposing philosophies have emerged which offer starkly different visions of what lies ahead. On one side stand the Doomers, a pessimistic bunch who see catastrophe on the horizon and believe society is circling the drain. Arrayed against them are the Accelerationists, who stubbornly cling to optimism and have faith that technology and innovation can overcome any challenges ahead. This fundamental divide is playing out in all facets of society, but perhaps most vividly in our educational institutions, where our youth grapple with these competing worldviews as they contemplate their own prospects.


For the Doomers, the future is bleak. Climate change, resource depletion, economic instability, tribalism and nationalism have put humanity on a crash course, and the Doomers see no realistic way to swerve out of its destructive path. In their philosophy, civilization has sown the seeds of its own collapse, and the reckoning is at hand. The concept of 'progress' itself is a fiction, and trying to perpetuate it is merely postponing the inevitable. Though their views may seem cynical, the Doomers believe it is better to acknowledge our plight and brace for impact, rather than chasing false hope. 


In contrast, the Accelerationists assert that the only way out is through. While civilizational challenges abound, technology and innovation can theoretically overcome them all. Accelerationists believe that the exponential growth of computing power, biotechnology, AI and more will unlock solutions to climate change, enable space travel, and radically transform society for the better. But to unlock this potential, humanity cannot shy away from its quest for progress - we must charge forward, accelerating technological and social advancement. Perfection may be unachievable, but profound improvements are within reach if society mobilizes its resources. Where the Doomers see a cliff's edge, the Accelerationists see open road.


Our youth are torn between these extremes. As they witness political rancor, societal injustice, environmental degradation, and an unmoored economy, it can feel as if the Doomers' pessimism is justified. But abstaining from ambition feels equally untenable for a generation told they can achieve anything through education and hard work. And so they find themselves caught in the crossfire of this philosophical tug-of-war, uncertain whether to strive boldly into the future or despairingly prepare for descension. 


Ultimately, both views hold strands of truth. Neither pure optimism nor unbridled pessimism serves us well. Yes, civilization faces existential threats, but cynicism alone will not surmount them. Our salvation lies somewhere between the Doomers' warnings and the Accelerationists' technological optimism, combining clear-eyed risk assessment with determined idealism. This more nuanced path is no doubt obscure and difficult. But the fate of humanity may well hang in the balance. And if we're to have any chance of forging a decent future from the flames of collapse, we must restore some semblance of wisdom, courage and measured hope.

This philosophical tug-of-war leaves many feeling disoriented and despairing. Without wisdom to light the way, purpose to motivate us, or hope to raise our eyes to the horizon, it is easy to detach and tune out the world's noise. This nihilism is understandable when bombarded by social media's negativity, politicians' tribalism, and news that spotlights society's bleakest happenings. 

Yet we must be careful not to overcorrect by completely abandoning engagement and ambition. Total apathy serves no one. While the path forward is uncertain, we are not helpless - we can still choose how to respond to this turbulence. Cynicism may feel temporarily comforting, but real courage lies in acknowledging harsh realities while retaining faith that progress is possible.  

With empathy, vigilance and moral imagination, we can build systems and structures that work for the many, not the few. Though the hour is late, a better society remains within reach if we collectively choose to stretch for it. This begins by rediscovering purpose, wisdom and hope so that we illuminate the darkness rather than retreat into it. For where there is light, even a flicker, there is the possibility of dawn.

Friday, February 23, 2024

The Real Job of Educators: Helping Kids Learn to Read

Our schools are being slowly choked to death by the insidious creep of administrative bloat. Layers upon layers of managers, consultants, and assorted bureaucratic barnacles have attached themselves to the education system, siphoning away precious resources that should be going to the classroom.

This army of educrats, as one might call them, fills their days with meetings, workshops, training sessions, and anything else they can imagine that justifies their existence. Actual teaching and learning seem like distant afterthoughts, so consumed are they with the latest fads and jargon about "learning outcomes," "stakeholder engagement," and other empty corporate-speak.

Meanwhile, teachers struggle with overcrowded classrooms, insufficient supplies, and mountains of paperwork. But the educrats are too busy admiring their own slick presentations and clever euphemisms to notice. Or perhaps they simply don't care, since they continue drawing their inflated salaries regardless of whether Johnny can read.

This is what happens when a system loses its why, its purpose. As the great Simon Sinek says, when there is no trust, no higher cause animating people's actions, you get all kinds of dysfunctional behavior - lying, faking, hiding. That is on full display in our education bureaucracy today.

What should be sacred - teaching children - has become secondary to keeping up appearances and preserving petty fiefdoms. True leadership and deep reform are needed to clear out the smothering layers of decrepit bureaucracy from our schools. We must remember that the purpose of education is not to serve bloated administrations and their delusions of usefulness - it is to nurture children's minds and prepare them for life. Until we get back to that guiding north star, our schools will remain lost in a bureaucratic wilderness.

The educators and administrators who benefit from and perpetuate bureaucratic bloat are masters at rationalizing the status quo. When challenged on wasteful spending or ineffective programs, they are quick to toss out excuses: 

"We don't have the resources." 

"It's too complicated to change."  

"We're too busy as it is."

"We have to follow district/state/federal policies."

There is always some justification for why things can't be improved or done differently. Meaningful self-examination is avoided at all costs - it might reveal that huge swaths of these administrative empires are redundant or useless. 

You'll almost never hear them take a step back and do a thoughtful cost-benefit analysis: Which programs are truly helping kids versus just keeping bureaucrats employed? What's the return on investment for our latest consultant or training scheme? They don't want to ask those questions because they know the answer will demand real change.  

It's easier for them to hide behind excuses and divert blame. They are afraid to challenge the status quo because it props up their power and salaries. Kids' education takes a backseat to the self-interest of bloated bureaucracies and their refusal to reform. Until this culture changes, the excuses and rationalizations will continue while schools languish.

Thursday, February 22, 2024

Create Your Own Endless Adventure: How AI Helps Kids Write Interactive Stories

Generating a Choose Your Own Adventure Story with LLM Anthropic AI

A Trailer For Sisters of Battle 100% Using Generative AI! 
The Future of CYOA and Intertainment? 

Introduction: 

For generations, choose your own adventure stories have captivated young readers, letting them craft exciting tales that unfold in countless directions. Now, AI is bringing these ever-branching narratives into the 21st century. With the help of natural language processing (NLP), kids can work with smart assistants to write interactive stories that truly never end. 

This innovative approach keeps young writers engaged, sharpening their skills while unleashing their creativity. In this article, we'll explore how AI empowers children to become co-authors of elaborate, customizable narratives. You’ll learn step-by-step how to brainstorm fascinating plot points, develop complex characters, and spin a story web that’s unique each time. With an AI assistant by their side, the possibilities are infinite for the next generation of choose your own adventure authors. So let’s dive in and start writing the never-ending story!

The focus on choose your own adventure stories, AI, NLP and kids in the title and intro aims to attract readers and optimize for relevant search terms. The intro summarizes the key points covered in the article for SEO as well. Let me know if you would like me to modify or expand the introduction in any way!

 Here are some detailed steps a child could follow to create an ongoing, real-time "choose your own adventure" story using Anthropic's AI assistant Claude:

1. Come up with an initial story premise or setting. Think about the main character, location, time period, etc.

2. Ask Claude to suggest 2-3 possible opening scenes that introduce the character and setting. For example, "Claude, can you suggest a few opening scenes for a story about a girl exploring a mysterious old house?"

Here are a few potential opening scenes to kick off a "choose your own adventure" story about a 4th-grade girl exploring a haunted house:
1. Jenny pedaled her bike up the twisty path leading to the abandoned Victorian mansion on the edge of town. Rumors said it was haunted, and she was determined to find out if it was true. As she walked her bike up the creaky steps to the front porch, she felt a cold wind blow past, giving her goosebumps. Jenny took a deep breath and pushed open the large front door.

2. "I bet you won't go inside the old Waterson house," taunted Jenny's friend Sam. "No one's lived there for years and they say it's full of ghosts." Jenny hesitated. She didn't really believe in ghosts...but what if they turned out to be real? "I'm not scared of any old house," declared Jenny. "Wait here if you want. I'll prove it's not haunted." Grabbing her flashlight, she marched up the weedy front walk. The door creaked open with a moan. Cobwebs brushed Jenny's face as she stepped inside. 

3. The kids at school dared each other to touch the rusty front gates of the looming Riddell House. No one knew what had happened to the family that used to live there, but rumor said they just vanished one night. Jenny reached out her hand, heart racing. The metal gate was cold under her fingers. Emboldened, she gave it a push and the gate swung open with a screech. Gravel crunched under Jenny's feet as she slowly walked up the path to the abandoned house. She imagined she saw a curtain move in an upstairs window. Taking a gulp, she reached for the front door handle...
3. Choose your favorite opening scene and ask Claude to generate 2-3 possible paths forward from there. For example, "Let's go with the scene where she finds a secret door. What are some options for what could happenext?"

4. Keep choosing which path to take, and asking Claude to generate more options moving forward. Make sure to vary the choices - some exciting, some mysterious, some funny.

5. Whenever you reach a good stopping point for a chapter, ask Claude to summarize the story so far. This helps keep the narrative coherent.

6. When you want input from friends, stop and copy/paste the story into a shared doc. Ask them to add their ideas for what should happen next.

7. Take your friends' suggestions and continue the story with Claude, occasionally checking back in with them for more input.

8. If you get stuck, ask Claude for help! For example, "I'm not sure what should happen next. Can you give me some ideas?"

9. When you reach an ending, ask Claude to generate a short epilogue to wrap things up.

10. Work together to come up with a title and fun cover art.

11. If you want to continue the story later, save your progress! You can pick up where you left off next time.

The key is leveraging Claude's creativity and your friends' ideas to collaboratively build an engaging, variable story! Adjust the steps as needed to suit your group's style. Most importantly, have fun!

Here is a "Choose Your Own Adventure" story starter worksheet that students can fill out to kickstart the creative process with an AI assistant:

Choose Your Own Adventure Story Starter

Main character name: 

Main character description (age, personality traits, etc.):


Setting (time and place):


Other characters:


Opening scene (a few sentences to introduce the character and setting):


Possible first plot points (2-3 ideas for what could happen first to set events in motion):

1.

2. 

3.

Potential challenges (things that could go wrong or obstacles for the character):

-

-


Ideas for possible endings (happy, sad, funny, scary etc.):


-

-

This worksheet provides a template to get started. Students can follow the steps we discussed to choose a starting point and beginning branching the story from there. They can always ask the AI for help filling in any sections they get stuck on. The key is to let their creativity run wild!

How AI Will Revolutionize Learning for Students with Special Needs

Top 10 AI Tools to Help Students with Dyslexia and Learning Disabilities 

Introduction: 

For students with learning disabilities like dyslexia and dyscalculia, education has often been a struggle. But rapid advancements in artificial intelligence and adaptive learning technology promise to remove learning barriers and revolutionize special education.

New AI applications are emerging that provide customized support based on each student's strengths, weaknesses, and preferences. From reading assistants that help decode text in real-time, to writing aids that improve composition skills as students type, to personalized tutors that target knowledge gaps - AI is making learning accessible, effective, and engaging.

These AI tools will enable immediate interventions, providing the equivalent of tier 2 and tier 3 supports that close achievement gaps for students with special needs. By tailoring instruction and feedback in real-time, AI promises to deliver the two sigma effect, dramatically accelerating learning. 

As the technology continues to develop, AI is poised to minimize the need for one-size-fits-all accommodations and empower students with special needs to thrive academically and beyond. The future of special education is bright as AI transforms learning into an adaptive, individualized experience for all students.

1. AI reading assistants that can read text aloud with natural inflection and help explain complex passages. This could make reading much easier for those with dyslexia.

2. AI writing aids that can help suggest vocabulary, grammar, and sentence structure as people type to improve writing skills.

3. Automated captioning and subtitles for videos, podcasts, and more. This makes content more accessible for people who process written information slowly.

4. Voice recognition software that can take dictation and transcribe speech to text quickly and accurately. This helps dyslexics communicate their thoughts in writing.

5. Math tutors that use AI to understand where students are struggling with dyscalculia and provide customized lessons and practice.

6. Apps that monitor reading progress over time and adjust difficulty levels accordingly to improve reading skills.

7. Intelligent keyboards that suggest correct spellings and grammar as people type to help improve writing.

8. Synthetic speech software that reads aloud written materials in a perfectly enunciated, understandable way for better comprehension.

9. Personalized recommendations for books, articles, and other materials based on reading level, interests, and needs.

10. Games and exercises that use AI to make learning math facts, arithmetic, and other concepts more engaging and adaptable to different learning needs.

The future looks bright for applying AI in ways that minimize learning barriers and allow people with dyslexia, dyscalculia, and other challenges to thrive academically and beyond. The possibilities are endless!

1. AI Reading Assistants

Advances in natural language processing will enable AI reading assistants to read text aloud to students with dyslexia with proper inflection, as well as explain complex passages to improve reading comprehension. These personalized reading aids will level the playing field in the classroom.

2. AI Writing Aids 

Writing aids powered by AI will provide real-time feedback on vocabulary, grammar, and sentence structure to students as they write. This will allow students with learning disabilities to improve their writing abilities and take the frustration out of writing.

3. Automated Captioning 

Automated captioning by AI will provide subtitles for classroom videos, podcasts, and more visual media. This captioning will make content more accessible to students who have difficulty processing written language.

4. AI Dictation Software

Voice recognition technology continues to improve and will soon allow students to easily dictate their thoughts which will then be transcribed quickly and accurately by an AI assistant. This helps learning disabled students better communicate through writing.

5. AI Math Tutors

Students struggling with dyscalculia will benefit from 1-on-1 tutoring sessions with an AI tutor. The AI will be able to recognize learning gaps and provide customized lessons in math to improve skills.

6. Adaptive Reading Apps

Apps that use AI to monitor a student's reading progress and adjust text difficulty accordingly will allow for data-driven improvement of reading skills for those with dyslexia. The apps make reading practice more accessible and effective.

7. Intelligent Writing Keyboards 

Enabled with AI, keyboards will provide real-time suggestions for correct spellings, grammar usage, and word choice as students type. This feedback loop when writing helps minimize mistakes and build skills.

8. Synthetic Speech Software

Text-to-speech software continues to improve, soon allowing written materials to be read aloud to students in a perfectly enunciated way. This improves comprehension of classroom materials for those with reading disabilities.

9. Personalized Recommendations 

AI programs will be able to analyze reading levels, interests, and academic needs to provide personalized recommendations for books, articles, and other materials for each student. This will enhance engagement and learning.

10. Adaptive Learning Games

Interactive games that leverage AI will be able to adapt in real-time to a student's needs, making learning math facts and concepts more engaging. This benefits students with dyscalculia and other learning needs.

Sunday, February 18, 2024

STEAM LESSON: Designing Skateboards Using the Stanford Design Process

Dear Students,

The Olympic Committee is excited to announce the Marathon Pour Tous, a "Marathon For All" event that will be held at the 2024 Paris Olympics. This event will consist of two races along the 2024 Paris Marathon route, and we are inviting students from all over the world to design a skateboard that can be used in this event.

We are looking for skateboards that are innovative, efficient, and safe. The skateboards should be able to handle the long distances of the marathon, as well as the various obstacles that the riders may encounter along the way.

We encourage you to use the Stanford design process to design your skateboards. This process consists of five steps:

1. 1. Empathize:

Understand the needs of the users. In this case, the users are the skateboarders who will be participating in the Marathon Pour Tous. What are their needs? What kind of skateboard would be best for them?

2. 2. Define:

Define the problem that you are trying to solve. In this case, the problem is to design a skateboard that is suitable for the Marathon Pour Tous. What are the specific requirements of the skateboard?

3. 3. Ideate:

Generate a variety of possible solutions to the problem. Brainstorm different ideas for skateboards that could be used in the Marathon Pour Tous.

4. 4. Prototype:

Build a prototype of your skateboard. This will allow you to test your design and make sure that it works properly.

5. 5. Test:

Test your prototype with skateboarders to get feedback. This will help you to improve your design and make sure that it is ready for the Marathon Pour Tous.

We are excited to see what kind of skateboards you come up with. We believe that this is a great opportunity for students to use their STEM skills to make a real difference in the world.

Sincerely,

The 2024 Paris Olympic Committee

Here are some additional tips for designing a skateboard for the Marathon Pour Tous:

Consider the materials that you will use. The skateboard needs to be lightweight and durable.

Think about the shape of the skateboard. The shape should be aerodynamic and comfortable for the rider.

Pay attention to the wheels. The wheels should be large and smooth, so that they can roll easily over rough surfaces.

Make sure that the skateboard is safe. The skateboard rider should have a braking system and other safety features to protect the rider.

We are confident that you can design a skateboard that is perfect for the Marathon Pour Tous. We look forward to seeing your designs!

Lesson 1 - Introduction to Skateboard Design

Objectives:

- Students will learn the parts of a skateboard and how they function together as a system.

- Students will be introduced to the Stanford design process.

Materials:

- Skateboard parts samples - decks, trucks, wheels, bearings, hardware

- Posters of the Stanford design process

Activities:

- Teacher leads discussion on skateboard parts using samples. Introduce key vocabulary - deck, trucks, wheels, bearings, hardware. (10 mins)

- Students disassemble and reassemble a skateboard in teams to understand how the parts connect. (10 mins)

- Introduce the Stanford design process - Empathize, Define, Ideate, Prototype, Test. (10 mins)

Lesson 2 - Empathize and Define the Problem

Objectives:

- Students will interview skateboarders to understand user needs.

- Students will define the design problem and criteria.

Materials:

- Skateboarding videos, magazines, websites for research

- Paper and pens for interview notes

Activities:

- Students research skateboarding culture and interview skateboarders to empathize with users. (15 mins)

- Class discussion on user needs and pain points. (10 mins)

- Teams outline the design problem and criteria for a marathon skateboard. (10 mins)

Lesson 3 - Ideate and Prototype

Objectives:

- Students will brainstorm and sketch design ideas.

- Students will build initial prototypes.

Materials:

- Blank paper, pens, markers

- Cardboard, tape, scissors and other materials for building

Activities:

- Teams brainstorm and sketch ideas for marathon skateboards. (15 mins)

- Teams select an idea and build an initial cardboard prototype. (15 mins)

- Teams present prototypes and receive peer feedback. (10 mins)

Lesson 4 - Test and Refine

Objectives:

- Students will test prototypes and gather performance data.

- Students will use feedback to refine designs.

Materials:

- Prototypes from previous lesson

- Tools for modifying - tape, scissors, markers etc.

Activities:

- Teams test prototypes through simulated runs and track performance. (15 mins)

- Teams interview users for feedback to guide modifications. (10 mins)

- Teams refine prototypes based on testing and feedback. (15 mins)

Skateboarding Takes Center Stage at the Tokyo Olympics

The 2020 Summer Olympics in Tokyo, Japan featured the debut of skateboarding as an official Olympic sport. After decades of influencing popular culture and creative expression, skateboarding finally took its place on the biggest athletic stage in the world. The introduction of skateboarding to the Olympic program marks a major milestone that legitimizes skate culture while expanding the Games to new audiences.

Skateboarding is divided into two disciplines for Olympic competition - park and street. Park skateboarding, also known as transition skateboarding, takes place on a course featuring a variety of curved ramps, dips, and bowls. Riders perform tricks by utilizing centrifugal force to careen up and down the curves and edges. In contrast, street skateboarding is done on a straight course meant to simulate objects found in an urban environment. Street skaters will grind handrails, slide over benches, jump down stairs, and flip their boards over gaps.

Both skateboarding disciplines made their Olympic debut in Tokyo with men's and women's park and street competitions. The skateboarding events were held at Ariake Urban Sports Park, which was specially built to host Tokyo's urban-based sports like BMX freestyle, 3x3 basketball, and sport climbing. The skate park's massive concrete surfaces gave Olympian skaters ample room to put together creative lines and wow judges with their best tricks.

The 80 Olympic skateboarders were selected through qualification systems administered by World Skate, skateboarding's international governing body. Skaters earned points by competing at events around the world, and the top ranked skaters in each discipline were awarded coveted spots at the Olympics. Big names like Sky Brown and Nyjah Huston represented their home countries, while amateurs from non-skateboarding strongholds like the Philippines and South Africa introduced the sport to new regions.

Ultimately, Japan and Brazil dominated the inaugural skateboarding competition. In a surprise upset, 12-year-old Kokona Hiraki of Japan took home silver behind Japan's Sakura Yosozumi in the women's park event. Japan also nabbed gold and bronze medals in men's street courtesy of Yuto Horigome and Sora Shirai. Not to be outdone, Brazil clinched the gold in men's park with Pedro Barros and two silvers from Rayssa Leal and Kelvin Hoefler in women's and men's street respectively. 

The skateboarding competition yielded some of the Olympics' most riveting moments. Sporting a helmet decorated with Anime characters, Brown crashed hard in women's park but recovered to land an awe-inspiring 540 wedged between the ramps. Jagger Eaton of Team USA whipped out his iPhone mid-run to take an ecstatic selfie after securing bronze in men's street. And 13-year-old Leal, the youngest athlete in Tokyo, endeared herself to fans worldwide with her infectious zeal while zooming around the course.

Beyond the actual competitions, skateboarding's Olympic debut also highlighted the activity's vibrant culture. Many Olympians sported their signature styles adorned with tattoos, jewelry, and dyed hair. Team Brazil paid homage to their street skating roots by wearing coordinated basketball jerseys on the podium. Skate apparel brands like Vans got in on the action, providing custom footwear for Olympians to rock while they shredded.  

In the end, skateboarding's introduction to the Olympics was a resounding success. Millions tuned in to watch skaters ply their athletic artistry on the world stage. Young fans were inspired to pick up a board for the first time, energizing skateboarding communities from Tokyo to São Paulo. As Paris prepares to host skateboarding again in 2024, the Olympics have ensured generations of skaters will continue carving out the sport's unique niche in the athletic pantheon.
Here is a comprehensive list of common skating lingo and words related to skate culture:

Skating Lingo

- Ollie - Popping the tail of the board to lift off the ground
- Kickflip - Flicking the board to make it rotate 360 degrees on the x-axis
- Heelflip - Flicking the board with your heel to rotate 360 degrees on the x-axis
- 360 Flip - A 360 degree spin combined with a kickflip
- Impossible - A 360 degree flip while turning the board 90 degrees
- Hardflip - A frontside pop shuvit combined with a kickflip
- Tre Flip - A 360 flip combined with a backside 180
- Varial Flip - Flicking off the side of the board to make it rotate on the z-axis
- Primo - Landing with all four wheels touching the ground
- G-Turn - Sweeping frontside turn on the nose or tail
- Nollie - An Ollie performed from the nose instead of the tail
- Fakie - Riding backwards
- Switch - Riding with the opposite foot forward
- Manual - Balancing on two wheels
- Nosegrind - Sliding along a ledge or rail on the underside of the nose
- 5-0 Grind - Sliding with both trucks grinding on the edge
- Crooked Grind - Grinding only on the back truck
- Lipslide - Sliding along the edge with the underside of the deck
- Bluntslide - Sliding along the edge balanced on the trucks


Skate Culture Words

- Shred - To skate with intensity and skill
- Stomp - To land a trick solidly
- Bail - To crash or fall off the board
- Eat it - To fall hard off the board
- Yard Sale - Falling and spilling gear everywhere
- Sketchy - Unsafe, out of control
- Mob - To skate a location with a group
- Flow - Smooth, seamless skating style
- Gnarly - Extremely difficult trick or spot
- Stoked - Extremely excited and hyped
- Core - True, dedicated skaters
- Poser - A skater who pretends to be something they're not
- Sponsored - Supported by skate companies with gear and money
- DIY - Building your own skate spot
- Vert - Vertical ramp or halfpipe skating
- Hip - A funnel-shaped transition on a mini ramp

The Forgotten Foundations: Listening, Speaking and the Science of Reading

The Science of Reading: Reclaiming the Lost Art of Listening and Speaking

Abstract

The science of reading has long focused on how children learn to decode text, at the expense of listening and speaking skills. Yet research shows that proficient listening and speaking abilities, collectively known as oracy, form the critical foundation for reading comprehension and academic success. This paper reviews evidence on the vital role of oracy in literacy development and synthesizes best practices for systematically cultivating listening and speaking skills from an early age. It argues that restoring oracy as the cornerstone of language arts curricula can significantly improve reading outcomes and reduce achievement gaps. The paper concludes by proposing specific policy and pedagogical changes to reintegrate the lost arts of listening and speaking into primary education.

Introduction

The science of reading has advanced tremendously over the past 50 years, yielding robust insights into how children learn to read. Drawing from cognitive psychology, linguistics and education research, we now have a detailed picture of the component skills involved in proficient reading, and strong evidence for instructional methods that effectively teach these skills (Rayner et al., 2001; Snow & Juel, 2005). However, amid the focus on decoding, phonics and comprehension strategies, a critical aspect of literacy development has been relatively overlooked in reading research and instruction: listening and speaking abilities, known collectively as oracy.

Oracy provides the fundamental language basis for literacy development. To read proficiently, children must first possess solid oral language skills, including a rich vocabulary, grasp of syntax and grammar, and verbal reasoning ability (Sticht, 2016). Yet, while research has firmly established the importance of oracy to reading achievement, policies and curricula seldom reflect this connection. The emphasis remains heavily skewed toward reading skills, while presuming students have adequate listening and speaking abilities. This imbalance reflects a lost art of oracy in modern education.

The marginalization of oracy in literacy curricula is concerning for several reasons. First, declining oracy skills may be contributing to stagnant reading achievement, particularly for disadvantaged students. Second, diminished emphasis on speaking robs students of opportunities to think critically and develop ideas collaboratively. Finally, lack of listening instruction forfeits a proven means of improving comprehension. This paper synthesizes research on the role of oracy in literacy development and proposes changes to restore listening and speaking as cornerstones of reading instruction.

The Vital Role of Oracy in Literacy Development

The Simple View of Reading conceptualizes reading comprehension as the product of decoding skills and linguistic comprehension, which depends heavily on oral language (Gough & Tunmer, 1986; Hoover & Gough, 1990). Specifically, oral language provides the foundation for vocabulary knowledge, listening comprehension, syntactical skills, verbal reasoning and knowledge of pragmatics that enable understanding of text (Language and Reading Research Consortium, 2015).

Evidence for the critical role of oral language in reading comes from studies showing strong associations between preschool oral language abilities and later literacy achievement. Language skills like vocabulary size, grammatical knowledge and narrative abilities in preschool predict growth in reading comprehension and writing skills years later (Dickinson & McCabe, 2001; Roth, Speece & Cooper, 2002). Longitudinal studies also reveal that difficulties in oral language foreshadow future reading problems. Children with language delays or disabilities in preschool are far more likely to develop reading disabilities compared to peers (Snowling, Bishop & Stothard, 2000).

Oral language influences reading development because listening comprehension provides the scaffolding for understanding text. Strong listening skills enable children to utilize contextual and syntactic cues to decipher meaning from language. This linguistic comprehension then transfers to making sense of print when formal reading instruction begins. Students with language weaknesses struggle to use context to infer meaning when listening, and in turn have difficulty generating inferences critical for text comprehension. Moreover, deficits in vocabulary and background knowledge resulting from impoverished language environments further hinder reading progress (Walker, Greenwood, Hart & Carta, 1994).

In summary, oral language constitutes a vital foundation for building literacy skills. Well-developed listening and speaking abilities provide the bedrock for vocabulary growth, syntactic knowledge, verbal reasoning and grasping meaning from language that allow written words to make sense. Given the centrality of oracy to reading proficiency, you would expect curricula to devote substantial time to intentionally cultivating listening and speaking skills. Unfortunately, this is often not the case.

The Lost Art of Oracy in Modern Classrooms

Walk into many elementary classrooms today and you will see students spending the bulk of their day engaged in reading and writing activities. Time devoted explicitly to listening and speaking is strikingly minimal by comparison. The marginalization of oracy is also reflected in literacy curricula and assessments. The thrust of standards and tests is decoding skills, strategies for comprehending written passages, and writing mechanics. Speaking and listening are relegated to an afterthought, if addressed at all.

How did oracy become a lost art in reading instruction? The roots trace back to the influence of behaviorism in education, focus on discrete reading skills in response to literacy crises, and continued academic stratification by socioeconomic status (Mercer, Warwick & Ahmed, 2017). Teaching the mechanics of reading eclipsed concern for holistic language development. The persisting inequities in literacy outcomes, particularly along racial/ethnic lines, further concentrated remediation efforts on basic skills instruction. This entrenched a fragmented view of literacy as simply the sum of parts like phonics and comprehension techniques, rather than a complex process rooted in language.

Another contributor is an underappreciation of how much foundational language development occurs at home before children enter school. Families with abundant economic and educational resources engage young children in rich conversational interactions that build essential language skills. They read aloud to children, use advanced vocabulary, and model syntactically-complex language. These early experiences scaffold literacy learning long before formal schooling begins (Rowe, 2012). Children raised in disadvantaged environments often lack these early language interactions, and thus begin school at a substantial oral language deficit. Yet education policies focus narrowly on school-based reading instruction starting in kindergarten, rather than systemic language development from infancy.

Finally, increased use of digital devices has radically altered how children engage with language, further crowding out opportunities for rich interpersonal talk. Screen media expose children to far more passive as opposed to interactive language. Even when caregivers co-view with children, televisions shows elicit less adult-child verbal engagement compared to book reading (Mendelsohn et al., 2001). Passive screen viewing surpassing social interaction from an early age impedes language development essential for literacy.

In summary, the diminished role of oral language instruction in modern classrooms reflects a confluence of factors: legacy of behaviorist approaches to literacy, inequitable early language exposure, and effects of pervasive screen media. The upshot is an imbalanced curriculum that focuses intensely on reading mechanics while failing to lay the oral language foundation needed to make sense of text. The overlooked importance of oracy helps explain a central dilemma in literacy education - why intensive reading remediation frequently fails to yield lasting comprehension gains, particularly for disadvantaged students.

Evidence-Based Practices for Developing Oracy

Given the vital role of oral language in literacy development, what specific listening and speaking skills should schools intentionally cultivate? Research points to key interrelated components of oracy that provide a scaffold for reading comprehension: vocabulary knowledge, narrative abilities, grammatical skills, and verbal reasoning aptitudes (Hulit, Howard & Fahey, 2010). Here we review evidence-based practices for systematically building these foundational oracy skills.

Vocabulary: Robust vocabulary knowledge is one of the strongest predictors of reading achievement because it facilitates making meaning from text (Cain & Oakhill, 2011). Effective methods for expanding students' vocabulary include: frequent read alouds exposing children to new words in context; direct instruction of word meanings; requiring students to use new words in context; and teaching word-learning strategies like using contextual cues, morphology and etymology (Senechal, Oulette & Rodney, 2006; Wright & Neuman, 2013). Educators must also be mindful of pervasive vocabulary gaps along socioeconomic lines, using methods to identify and mitigate these inequities early (Gilkerson et al., 2017).

Narrative Skills: Learning to comprehend and construct meaningful narratives provides a bridge between oral language and understanding text (Clarke et al., 2010). Strong narrators learn to sequence events coherently, convey causal connections between plot points, and integrate complex syntax and vocabulary to build meaning. Effective approaches include modeling story grammar and complex language through read alouds; eliciting personal stories from students focused on temporal ordering; and teaching text structure strategies that strengthen reading comprehension (Spencer & Slocum, 2010).

Grammar: Knowing the syntactic rules of language facilitates both expression and comprehension. Yet grammatical instruction has declined, contributing to deficits in students' language knowledge (NCEE, 1983). Renewed explicit emphasis on grammar is needed, including morphologic analysis of word parts, the functions of phrases and clauses, and conventions for complex sentence formation (Bardovi-Harlig, 1992; NIPD, 2008). Syntax should be taught using engaging texts that illustrate grammatical principles in action.

Verbal Reasoning: Higher-order language skills like making inferences, understanding metaphors, and grasping intent and purpose underlie the transition from basic language comprehension to interpreting written text (Van Kleeck, 2008). Methods like asking open-ended questions, modeling think-alouds, and engaging students in collaborative dialogue build verbal reasoning essential for academically-oriented language use (Mercer & Dawes, 2014).

In addition to targeting these specific components, certain instructional principles maximally support oracy development: student-centered dialogue emphasizing conceptual language rather than display questions; cognitively-challenging talk requiring reasoning and analysis; scaffolding complex language through modeling; and building on students' cultural/linguistic backgrounds (Cazden, 2001; Flynn, 2016). Integrating these evidence-based practices systematically into curriculum from preschool onward can significantly strengthen oral language abilities that scaffold reading comprehension.

Conclusion: Restoring Oracy in Literacy Instruction

The science of reading has revealed much about how children develop skills to decipher text. However, unraveling the written code is only half the equation. Proficient reading equally depends on the language foundation that allows decoded words to have meaning. This linguistic bedrock is built through the lost arts of listening and speaking that ought to be pillars of primary literacy education.

Several policy and pedagogical changes could restore oracy to its essential place in reading instruction:

1) Literacy curricula need to be rebalanced to devote comparable time to intentionally building speaking and listening skills. Language arts standards must also dedicate equal emphasis to oracy.

2) Teacher training and professional development must include best practices for developing oral language across domains like vocabulary, narrative discourse, grammar and verbal reasoning.

3) Schools need to screen for oral language skills starting in preschool and provide additional language intervention for at-risk children.

4) Family involvement initiatives should educate parents, starting prenatally, on the importance of rich adult-child oral interaction for literacy success.

5) Given that language gaps manifest early, we need to rethink the late onset of formal reading instruction in kindergarten. Earlier cultivation of speech, vocabulary and comprehension in preschool may level the playing field.

The science of language leaves no doubt that listening and speaking provide the developmental soil from which literacy blooms. Reading is not a culture-free skill that can be installed in children simply by drilling phonics and activating prior knowledge. Rather, learning to read and write depends on a foundation of oral language experience that too many children lack. Restoring oracy as an essential cornerstone of literacy education can help remedy skill gaps and realize the promise of universal reading proficiency. But first, we must rediscover and revitalize the lost arts of listening and speaking.

The Explosion of Alpha Students Disrupting Classrooms

The Explosion of Alpha Students Disrupting Classrooms

In many schools across the country, teachers are struggling with an increasing number of disruptive and overconfident students who seem to lack empathy, social skills, and the ability to follow basic classroom norms. These "alpha" students exhibit behaviors associated with the Dunning-Kruger effect - they overestimate their own abilities and competence, while being unable or unwilling to recognize their deficiencies.

While a degree of confidence can be positive, teachers report that these alpha students take it too far. They try to dominate classroom discussions, talk over others, refuse to listen to instructions, and seek to gain attention and control over their peers. Instead of coming to class ready to learn, these students come with the primary aim of being the "shot caller," the one who is in charge. Teachers describe chaotic scenes where these alpha students openly bully other students, disrupt lessons, and make it difficult for anyone else to learn.

Experts posit a few key factors that may be contributing to this phenomenon:

Lack of In-Person Socialization

After two years of remote learning during the pandemic, many students are lacking in-person social skills. They have spent minimal time interacting face-to-face with peers and learning how to make friends or work collaboratively. They have relied primarily on virtual connections. This lack of interpersonal practice and isolation from in-person peer dynamics may stunt their ability to pick up on social cues, compromise, and understand classroom social hierarchies. They may resort to aggressive behaviors in an attempt to assert social dominance.

Excess Screen Time

Related to remote learning, excessive screen time has been linked to issues with self-regulation, attention span, empathy, and aggression in adolescents. Most students today are spending hours each day staring at phones, tablets, computers, video games and TV. While useful in moderation, too much screen time can negatively impact social-emotional skills. Students then bring those poor skills to the classroom environment.

Unstructured Home Environments

For some students, homes may lack structure, rules, and enforcement of norms. Students from permissive households where they are not held accountable for bad behaviors often struggle to adjust to classrooms where rules matter. They are unaccustomed to conforming their conduct based on external expectations. These home dynamics, if too lax and permissive during childhood, fail to instill self-control and appropriate classroom behaviors in students.

Lack of Consequences

Exacerbating the issue, many teachers and administrators feel limited in their ability to apply consequences to alpha students who disrupt classrooms. Fearful of being accused of discriminating against students with diagnoses like ADHD or Oppositional Defiant Disorder, they tiptoe around discipline issues that would have been addressed firmly in previous eras. Students quickly pick up on this hesitancy and use it to their advantage. Their bad behaviors escalate as they push boundaries and realize there are few real penalties. They quickly take over classrooms without repercussions.

Limited Teacher Support

Finally, teachers often lack adequate support when dealing with these severe student behaviors. Oversized classrooms, inclusion of students with significant issues, and lack of classroom aides put teachers in incredibly demanding situations. Often there are a few intensely disruptive alpha students that make it nearly impossible for teachers to teach and well-behaved students to learn. Teaching today has become an extremely stressful profession, and many teachers do not feel they have the back-up they need from administrators when it comes to discipline issues.

Possible Solutions

To address the explosion of alpha students determined to disrupt classrooms and interfere with the learning environment, experts recommend taking a multi-pronged approach:

- Social-emotional learning: Integrate regular SEL lessons focused on developing empathy, cooperation, conflict resolution and relationship skills. Use role-playing activities to practice appropriate conduct.

- Positive reinforcement: Counter disruptive behaviors with a system of positive reinforcement. Use rewards, praise, special responsibilities and privileges to incentivize good conduct.

- Consequences: Establish clear consequences for violations of classroom rules and apply them consistently. Follow through on both rewards and consequences.

- Parent outreach: Maintain frequent contact with parents of struggling students. Enlist them as partners to encourage proper school behaviors. Provide suggestions for structures and routines at home.

- Staff training: Ensure teachers are trained in classroom management strategies tailored to today's students. Provide them support through teacher aides, counselors, school psychologists, and administrators.

- Relationship building: Teachers should dedicate one-on-one time to get to know disruptive students and understand underlying causes of their behavior. Building trust and rapport can lessen power struggles.

- Referrals: Implement systems for referring severely disruptive students to counseling, evaluations and services. Provide individualized plans and support for students with behavioral disorders.

- Alternatives: Develop alternative learning environments, like small group classes with higher staffing ratios, for students unwilling or unable to function in regular classrooms.

By utilizing a combination of these strategies, schools can minimize classroom disruptions and refocus energy on creating positive learning communities where all students have the opportunity to thrive. But it will require commitment and resources to provide staffing, services and environments that meet the challenges teachers face with today's alpha students.