Artificial Intelligence in Limestone

Prioritize ethical considerations, accountability, and transparency, in support of respectful human interaction, through responsible education, training, and usage.

Ensure access to AI is inclusive and equitable, minimizing biases, and supporting diverse learning needs.

Use AI to improve and innovate teaching and learning, and improve student achievement and well-being. Equip staff and students with the skills to use AI responsibly to enhance the educational environment.

Strive for continuous innovation and improvement through experimentation and learning, where feedback is used to refine and enhance AI use.

Protect personal and confidential data, comply with privacy laws, implement robust security measures, and clearly communicate AI use.

AI research dates to the 1950s, when British mathematician and computer scientist Alan Turing proposed the Turing Test in his paper entitled Computing Machinery and Intelligence[1]. The Turing Test is a method for measuring how well a machine can demonstrate behaviors and functions that are typical of human thinking, such as logical reasoning and problem-solving. The Turing Test involves an evaluator engaging in separate text-based conversations with two subjects: one participant is a human, and the other is a machine. A machine would pass the Turing Test if it responded in a way that convinced the evaluator it was the human — or at least made it difficult for the evaluator to reliably tell which participant was the machine.

Advancements in AI research have continued and we’ve seen breakthroughs in areas such as deep learning and natural language processing. These advances have led to AI becoming an integral part of various sectors, such as healthcare, finance, business, and education.

[1] Turing, A. M. (1950). Computing machinery and intelligence. Mind, 59(236), 433–460. https://doi.org/10.1093/mind/LIX.236.433

AI is embedded into various aspects of our society, impacting our daily lives and enhancing productivity in many ways. Examples of AI systems include voice assistants (like Siri or Alexa), recommendation systems (like those used by Amazon, Netflix, or YouTube), as well as facial recognition scanners at airports.

Over the years, AI technology has advanced at a fast pace and created new opportunities. Whether you realize it or not, many of the tools that we use daily are powered by AI technologies.

Generative AI (Gen AI) creates new content, such as text, music, and images, based on a user-provided prompt. For example, Gen AI tools such as Microsoft Copilot can be used for everyday tasks, including generating an e-mail draft, creating a trip itinerary based on interests of the user, or creating a recipe including only ingredients that are on hand, while also considering dietary restrictions.

Predictive AI, often used in recommendation systems, makes predictions by analyzing past data and inferring patterns. For example, it might predict a movie that a user would like to watch based on the movies that the user has previously viewed. Financial institutions apply predictive AI technologies to classify the risk of mortgage applicants and to identify fraudulent credit card transactions. Social media platforms even use AI facial recognition algorithms to identify people in media that is uploaded to the platform.

Reactive AI systems do not generate new content; instead, they respond to user prompts based on traditional algorithms. Examples of reactive AI systems include Siri and Alexa. If you ask Siri or Alexa, 'What is the weather today?' the system will provide the current weather. Similarly, if you ask Siri or Alexa to tell a joke, the system will simply select a joke from a pre-existing collection, rather than creating a new one in real-time. This is different from a generative AI tool, which might create a new joke and tailor it based on the user’s past interactions and preferences.

Generative AI can create various types of content, including text, music, videos, images, and even 3D models. When generating text, Gen AI typically uses a large language model (LLM) that is trained on extensive amounts of human generated content, including articles, websites, books, and social media posts. The AI system analyzes this data to identify patterns in how words are used, how they are commonly paired, and how sentences are structured. Using these patterns, the AI generates new content based on the user’s prompt. AI systems lack sentience – they generate content by analyzing and replicating patterns in the data that they have been trained on, without knowledge of the context or meaning of the content that they generate. A helpful comparison is a parrot: it can mimic human speech, but it doesn’t actually understand the words it’s saying. Similarly, AI produces responses without comprehending the meaning. AI tools can sometimes 'hallucinate,' meaning they may produce information that is false or misleading. It is important for users to verify AI generated content to make sure it is accurate and relevant.

AI enhances existing teaching and learning approaches through providing new opportunities for personalized and engaging learning experiences. Examples include:

• Personalized Learning: Gen AI can customize educational experiences based on student interests, strengths, progress, and areas for improvement. For example, AI tools can simplify complex topics by providing examples that are relevant to the student. Additionally, AI-powered platforms can design dynamic lessons that focus on the specific skills the student needs to develop.
• Promoting Creativity: Gen AI can assist students with creative tasks, including the creation of art, music, and multimedia presentations.
• Research Support: Gen AI can help students locate sources on a topic and summarize texts by outlining the key points.
• Organization of Ideas: Gen AI can assist students in creating outlines for their texts and generating mind maps that visualize the relationships between different ideas.
• Feedback and Revision: Generative AI provides students with real-time feedback to improve their writing, thereby fostering self-assessment and growth.
• Executive Functioning: Artificial intelligence tools can assist students in scheduling tasks, setting goals, and monitoring progress towards those goals.
• Studying Resources: Gen AI can be utilized to generate learning materials including flashcards, levelled texts, exemplars, and self-assessment quizzes.
• Interactive Discussions: Students can engage with AI assistants that simulate conversations with various characters including historical figures, literary characters, or experts in different fields. These tools provide students with engaging ways to explore new ideas, concepts, and perspectives.

A focus group of Grades K–12 Limestone educators is currently exploring how AI tools can be used in teaching to support student learning. Feedback from educators and students will be gathered to guide future steps for Limestone’s implementation of Gen AI. The focus group is currently piloting the following AI tools:

• Brisk: Brisk supports educators in developing lessons, differentiating instruction, providing feedback, adjusting text reading levels, offering guidance on student writing, and creating activities such as quizzes that align with provincial curriculum standards.
• Co-Pilot: Co-Pilot is an AI tool incorporated into Microsoft 365 that assists educators with various administrative tasks, including drafting newsletters, revising text, organizing schedules, designing engaging lessons, and creating educational materials for students.
• Magic School: Magic School supports educators in designing personalized and adaptive lessons, creating differentiated assessments and rubrics, and producing learning materials tailored to students' interests and needs.
• School AI: SchoolAI enables educators to create engaging learning environments called "spaces" for individual student interactions with AI. Through these "spaces", students are able to obtain real-time, personalized support about the topics that they are studying. Educators define objectives and success criteria to ensure curriculum aligned learning.
• Quizizz: Quizizz supports educators in creating interactive activities such as: quizzes, lessons, polls, and exit tickets, as well as, study aids like flashcards and vocabulary lists. These activities can be customized to meet student needs and aligns with provincial curriculum expectations. Educators can track student progress using the dashboard to identify next steps for teaching and learning.

AI-driven tools can support students in learning and using technology, as well as fostering student creativity. For instance, generative AI applications such as Microsoft Co-Pilot can generate HTML and CSS code to create a website. If a student starts their own business and wants to market their products, AI can help the student build a website for their business, even if they have limited knowledge of computer programming. Additional use cases for students to use Gen AI in their daily lives include:

• getting feedback on a resume or cover letter for a summer job.
• engaging in mock job interviews with AI and receiving real-time feedback.
• searching for jobs that match their skills, experiences, and interests.
• managing a personal budget, tracking expenses, and setting savings goals.
• organizing their time, setting reminders, and scheduling daily tasks.
• creating artwork and videos, composing music, and designing engaging presentations.

We live in an increasingly digitalized and connected world. It is more important than ever for our students to understand how computers work and the power that they offer to support everyday tasks and increase productivity.

Artificial intelligence has an impact across all sectors of society. In the banking industry, AI algorithms are employed to analyze transaction patterns and identify suspicious activities. In various trades, such as welding, AI-powered systems enhance efficiency, precision, and safety, thereby reducing defects. Furthermore, in healthcare, AI assists in identifying abnormalities in medical scans and supports doctors in making informed treatment decisions.

Technology is present in every sector, and most careers will involve using computers to some extent. By teaching students how to use AI, they will gain the knowledge and skills needed for future success, regardless of their chosen career path. Additionally, understanding how AI works provides students with the necessary background to make informed decisions about using this emerging technology responsibly and productively.

In Limestone, we prioritize ethical use, accessibility and equity, teaching and learning, enhancement and improvement, and protection and privacy in our use of Gen AI tools. The guidelines serve as a basis for guiding the use of GenAI tools in teaching and learning.

• Ethical Use: Prioritize ethical considerations, accountability, and transparency, in support of respectful human interaction, through responsible education, training, and usage.
• Accessibility and Equity: Ensure access to AI is inclusive and equitable, minimizing biases, and supporting diverse learning needs.
• Teaching and Learning: Use AI to improve and innovate teaching and learning and improve student achievement and well-being. Equip staff and students with the skills to use AI responsibly to enhance the educational environment.
• Enhancement and Improvement: Strive for continuous innovation and improvement through experimentation and learning where feedback is used to refine and enhance AI use.
• Protection and Privacy: Protect personal and confidential data, comply with privacy laws, implement robust security measures, and clearly communicate AI use.

In Limestone, we are committed to equipping our students with the digital literacy skills that they will need to be responsible and productive users of digital technologies, both at school and beyond. Students will learn how AI tools work, the capabilities and limitations of AI-powered tools, and strategies to critically assess the content that they produce for accuracy. Students will also learn about the risks associated with AI, such as deepfakes, false information, and oversharing of personal information. When students understand how AI works and how it learns from user-provided data, they can better assess the suitability of the information they input into and receive from AI tools.

Students can use AI as a tool to support and enrich their own original ideas and creativity in productive and responsible ways across subject areas. Strong digital literacy skills equip students with the ability to critically evaluate online information, stay safe online, and prepare them for the future.

The provincial curriculum introduces emerging technologies in elementary science, secondary geography, technological education, and computer studies courses.

The Ontario Ministry of Education’s transferable skills outline essential competencies for students to succeed in a digitalized and connected world. These competencies include:

• critical thinking and problem solving
• innovation, creativity, and entrepreneurship
• self-directed learning
• collaboration
• communication
• global citizenship and sustainability
• digital literacy

By teaching students how to use AI, we are helping them develop important skills for the future. AI is offering students opportunities to innovate and be creative in areas like entrepreneurship, global citizenship, and sustainability. Students can use their knowledge of digital technologies, like AI and coding, to create solutions for real-world problems that directly benefit their local communities.

At Limestone, AI implementation is approached by offering students opportunities to learn about emerging technologies while prioritizing relationships, collaboration, and the development of future-ready skills.

AI enhances personalized learning for students in various ways and complements existing teaching practices. Educators can use AI-powered tools to create interactive, personalized learning experiences. AI also supports student learning by providing instant support, such as by explaining concepts and answering questions in real-time. Additionally, many AI-powered tools feature assistive technologies, including text-to-speech, predictive text, and other accessibility functions.

AI-powered tools have the potential to enhance teaching and learning by supporting educators with tasks, such as:

• Crafting individualized feedback, while freeing up educator time for direct student support.
• Generating quizzes, worksheets, flashcards, learning activities, and projects that align with specific learning goals, success criteria, and the Ontario curriculum expectations.
• Personalizing learning, including designing lessons that align with student interests, providing targeted intervention based on student needs.
• Creating interactive activities such as simulations of science experiments and conversations with literary characters.
• Adjusting the reading level of texts to differentiate learning for students.
• Tailoring and adapting learning activities based on a student's progress.
• Serving as an effective starting point in the lesson planning process.

It is important to recognize that AI lacks essential human qualities like empathy, compassion, and critical thinking, underscoring that the use of AI-powered tools will not replace the central role that educators play in the lives of their students. AI can be a valuable resource for educators, assisting with administrative tasks and freeing up time for personalized student support.

Yes, many of the available AI-driven AI tools support French language queries and content generation, which makes these tools accessible for French as a Second Language educators and students.

Educators continue to follow the Ontario Ministry of Education's Growing Success guidelines to ensure assessment and evaluation practices are fair, transparent, equitable, and include triangulation to collect evidence of student learning from multiple sources including observations, conversations, and products.

Educators assess and use their professional judgment to identify the educational benefits of various digital tools, determining which ones to include in their classrooms and how to utilize them effectively for teaching and learning. AI use varies across disciplines. For instance, a science educator may employ AI-powered tools to create interactive simulations for teaching scientific concepts, while a history educator might use AI to simulate conversations with historical figures.

Students will have opportunities to use AI to support their learning. If students have any questions about the permitted use of AI in a specific class or assignment, they are encouraged to ask their teacher for guidance.

There are many frameworks for developing effective AI prompts. Here are some general tips:

• Be specific and include any relevant background information.​
• Specify the role that the AI should assume​.
• Specify the desired output of the response​, including the format.
• Give the AI tool information on the target audience and the desired tone of the response.​
• Specify any relevant restrictions or parameters.​
• AI prompting is an iterative process – if you don't get the desired response the first time, provide feedback and re-prompt.

Example Prompt: My child is in grade 7 and is learning about the particle theory of matter. Can you guide my child through a conversation about the particle theory of matter to help them understand this topic better? Please use simple language, real-life examples, and base the conversation on the Ontario Science and Technology curriculum. The conversation should guide my child through the discussion by prompting them with questions to help them think critically without directly giving them the answer. Please focus on using the particle theory to distinguish between pure substances and mixtures.

Consider a scenario where students are studying the topic of velocity, and one student is an enthusiastic soccer player. Microsoft CoPilot, a generative AI tool, can be utilized to explain the concept of velocity from the viewpoint of a soccer player. This example illustrates how AI can differentiate and customize learning according to student interests.

Microsoft. (2023). Microsoft Copilot [AI-powered assistant]. www.microsoft.com/copilot

Limestone is committed to protecting student privacy and data security. We ensure compliance with relevant legislation such as PIPEDA and MFIPPA. AI systems used in Limestone are selected with a focus on safety and privacy, and data is anonymized whenever possible.

Limestone has implemented guidelines for responsible AI use in the classroom. These guidelines highlight expectations about how AI can be used in ways that support student learning while prioritizing privacy and data security. Parents and guardians will be informed when AI tools are being used in the classroom.

Staff will engage in ongoing professional learning related to AI with a focus on effective AI use, building AI literacy, and expectations related to data privacy and security.

Most AI-powered tools have age restrictions. For example, ChatGPT is not intended for use by children under 13. Children aged 13 to 18 are required to obtain parental consent before using ChatGPT. Google’s Gemini requires that all users are 18 years of age and older.

Many popular apps such as Facebook, Instagram, and Snapchat have AI features embedded within the app. Parents and guardians are encouraged to be aware of the digital tools their students are using, the terms of use and privacy policies for the platforms and set expectations around how their student uses these apps. 

As with all technologies, it is important to maintain ongoing communication with your student about how they are using digital tools, how to stay safe online, as well as how technology can be used in safe, productive, and responsible ways. Parents are encouraged to support their students in thinking critically about content that they see online, being aware of what personal information they are sharing online and avoiding improper uses of AI tools (e.g., plagiarism, deepfakes, and cyberbullying).

In Limestone, we view AI as a tool that can be used in responsible and productive ways to enhance teaching and learning. AI does not replace the role of the classroom educator but rather supports the educator in offering students innovative and personalized interactive learning experiences. Al is powerful and presents many opportunities to transform education, but there are also limitations to the tools. Educators will continue to use their professional judgement to balance the use of AI-powered tools and evaluate the relevancy, accuracy, and suitability of all AI-generated content. Educators will ensure that AI generated content meets the needs of their learners and aligns with established learning goals, success criteria, and the Ontario curriculum. Some limitations of AI are:

Bias: AI can exemplify bias in the responses that it generates because it learns from existing data, which may already contain systemic biases. The AI output can amplify historical and societal stereotypes present in the training data. For example, if an AI-powered tool is asked to generate an image of a firefighter, it may produce a picture of a male firefighter. Since firefighting has historically been a male-dominated field, the training data can overrepresent men in this career area.

Hallucinations: AI generated information is not always accurate. AI tools can sometimes produce incorrect information, known as hallucinations, so it is important for users to critically evaluate the results.

It is important to remember that AI systems do not have thoughts, feelings, or intentions. They generate content without truly understanding its context or meaning. Therefore, users should carefully evaluate all AI-generated information to ensure it is accurate, relevant, and appropriate.

As we teach students how to use AI-powered tools, an emphasis will be placed on building students’ critical thinking and digital literacy skills, and equipping them with strategies to consider broader societal implications and ethical considerations associated with AI technologies. These skills will not only help students to use AI in ethical ways but will also support them in being able to identify misinformation in other mediums, such as videos and reels shared on TikTok or Snapchat or while evaluating bias in online news articles. AI-driven tools rely on the training data they are given, and they may produce output using information that is no longer current.

Students will engage in learning related to risks associated with AI, such as deepfakes, false information, and oversharing of personal information. When students understand how AI works and how it learns from user-provided data, they can better assess the suitability of the information they input into AI tools.

Students will learn how to use AI as a tool to support and enrich their own original ideas and creativity in productive and responsible ways across subject areas. Strong digital literacy skills equip students with the ability to critically evaluate online information, stay safe online, and prepare them for the future.

• Some AI tools will provide a list of sources that the AI used to generate its response. Students can visit the references to verify the content and ensure that the AI tool correctly interpreted the information.
• Students can ask the AI tool follow-up questions to dig deeper into the response and potentially identify ambiguities and inconsistencies in the generated content. For example, the response to a follow-up question might contradict a previous answer, which would be a red flag that it could be inaccurate.
• Students can verify AI’s claims by cross-checking AI generated content with information from primary sources.
• AI-powered tools should be used as a starting point and a learning partner, rather than being used for tasks without human oversight. Students should let their creativity shine and ensure that their use of AI is not inhibiting their original ideas and thinking. When students are using AI to research information, they should draw on their background knowledge and information that they already know about the topic as they evaluate the accuracy, relevancy and appropriateness of AI generated content.