Embedded Files
What We Learned Together
What We Learned Together
Over the course of this Opportunity Forecast, the first cohort of participants explored the opportunities, challenges, and future potential for AILPs from a variety of educational and professional perspectives. Through discussions, collaborative activities, and design challenges, our cohort generated thoughtful insights that both reinforced and extended the research presented throughout this OER.
The reflections below summarize some of the common themes that emerged across each activity that demonstrate our collective thinking and how each participant's ideas evolved throughout this learning experience.
Launchpad Discussion
Launchpad Discussion
Participants kicked off our OER on the ETEC522 Launchpad Post by sharing their initial reflections on what refined an AILP, how it differs from traditional AI tools, and where its greatest value lies across diverse learning contexts.
Summary of Themes
Summary of Themes
Coach, Not Search: Respondents consistently distinguished AILPs from standard chatbots or search engines. Instead of delivering direct answers, an AILP is envisioned as an interactive thinking partner that uses dialogue, probing questions, and scaffolding to guide learners toward their own conclusions.
Struggle vs. Speech: A central focus was on ensuring AI support does not eliminate the friction necessary for deep learning. Participants noted that providing answers too quickly risked creating dependency and sacrificing long-term comprehension for immediate task completion.
Safety Without Stigma: Many participants highlighted the unique value of AI's "unlimited patience." It provides a safe, low-stakes environment where learners can ask basic questions, make mistakes, and build confidence without judgment.
Realities & Barriers: While recognizing the potential of AILPs, participants also stressed that AI cannot replace human vulnerability and collaborative growth. Key concerns included data privacy, paywalls, teacher/educator workload, and the challenge of balancing AI support with classroom dynamics.
Participant Voices
Participant Voices
The reflections below illustrate key insights shared during the launchpad discussion:
"An AI Learning Partner functions much more like a collaborative coach... It maps out a path using hints and socratic dialogue to guide them to their own conclusions." — Erika
"AI provided a judgment-free space where I could ask as many questions as I needed.. [but] it cannot replace the growth that comes from being vulnerable with other people." — Cherri-Lynn
"I think an AI Learning Partner is not only there to answer questions, but to also guide you to the answer of the assignment through questions and dialogue." — Logan
"If AI provides answers too quickly, it risks reducing opportunites for thinking, processing, and reflection. Learning takes time..." — Denise
Overall Reflection
Overall Reflection
Across 35 participant comments, the cohort established that the true potential of AI in education isn't about automating answers, but instead about designing thoughtful, human-centered partnerships that support critical thinking, confidence, and genuine learning agency.
Reflection Activity
Reflection Activity
Summary of Themes
Summary of Themes
AI Learning Partners should foster metacognition and reflection: Across nearly every context, participants identified metacognition and reflection as the defining characteristics of an AILP. Rather than simply providing answers, particpants envisioned AI prompting learners to better explain their thinking, reflect on misconceptions, set goals, and support self-directed learning. Many viewed these reflective processes as more valuable than task completion or answer generation.
AI Learning Partners should foster metacognition and reflection: Across nearly every context, participants identified metacognition and reflection as the defining characteristics of an AILP. Rather than simply providing answers, particpants envisioned AI prompting learners to better explain their thinking, reflect on misconceptions, set goals, and support self-directed learning. Many viewed these reflective processes as more valuable than task completion or answer generation.
Personalization should support, not replace, teachers: While personalization was widely valued, participants also described AI as a tool that should augment teacher decision-making instead of replacing teacher expertise. Many highlighted the importance of maintaining relationships, professional judgment, and human guidance alongside AI-supported learning. Several participants also suggested that AI could free teachers to spend more time supporting learners with greater needs.
Implementation matters as much as the technology itself: Participants moved beyond discussing AI capabilities to ask practical implementation questions. They considered issues such as teacher oversight, privacy, classroom workflows, productive struggle, and when learners should seek teacher support instead of over-reliance on AI tools. Several participants questioned who determines an AI’s curriculum, values, and pedagogical approach.
AI Learning Partners have broad educational potential: Across K–12, post-secondary, language learning, corporate learning, and professional learning contexts, participants consistently prioritised learner agency, critical thinking, and lifelong learning over efficiency or automation. Many argued that the greatest value of AI Learning Partners lies not in providing faster answers, but in helping learners develop transferable thinking strategies that extend beyond a single task or subject.
Participant Voices
Participant Voices
The reflections below illustrate key insights shared during the market discussion:
"..the greatest value an AILP can bring to the classroom is teaching students how to become more self-directed learners..." — Jake (K-12)
“Who determines the AI’s curriculum, teaching style, pedagogical approach, and core values?… The implementation seems just as important as the technology itself.” — Denise Mouton (K–12)
“Personalization is only truly effective when learners understand how they learn, why they make progress, and what strategies will help them improve.” — Megan Frederick (K–12)
"... AI should encourage learners to plan, monitor, and evalute their own thinking..." — Manouchehr (Post-Secondary)
"... reflection involves exploring experiences, values, decisions, and possible pathways rather than finding a single correct answer." — Jennifer (Career Education)
".. personalized AI learning partner adapts scenarios, difficulty, and pacing in real-time..." — Erika (Corporate Learning)
Market Activity
Market Activity
In the Market Activity, participants chose the most and least suitable AILPs for their needs, then compared these two based on usability, accuracy, and functionality by asking both for the same output based on the same sources.
Summary of Themes
Summary of Themes
AI tools should be matched to the learning task: The most common theme was that there is no single "best" AILP. Instead, users recognized that different tools excel at different functions. Specifically, users noted that NotebookLM was stronger for source-grounded research and synthesis; Claude excelled at structured, pedagogically organized outputs; and Gemini was better suited for rapid, aesthetic content generation.
AI tools should be matched to the learning task: The most common theme was that there is no single "best" AILP. Instead, users recognized that different tools excel at different functions. Specifically, users noted that NotebookLM was stronger for source-grounded research and synthesis; Claude excelled at structured, pedagogically organized outputs; and Gemini was better suited for rapid, aesthetic content generation.
Usability strongly influences educational adoption: Beyond output quality, many participants discussed how interface design and workflow affected their impressions of each tool. Several participants acknowledged that usability shaped their perceptions of trust and usefulness, suggesting that successful AILPs must aim for pedagogical quality as well as an accessible user experience.
Usability strongly influences educational adoption: Beyond output quality, many participants discussed how interface design and workflow affected their impressions of each tool. Several participants acknowledged that usability shaped their perceptions of trust and usefulness, suggesting that successful AILPs must aim for pedagogical quality as well as an accessible user experience.
AI should support learning rather than replace thinking: A recurring educational theme was that AI Learning Partners should scaffold learning rather than provide answers. Several reflections also emphasized that students should critically evaluate AI-generated information instead of accepting it uncritically, reinforcing the teacher's continuing role in facilitating learning.
AI should support learning rather than replace thinking: A recurring educational theme was that AI Learning Partners should scaffold learning rather than provide answers. Several reflections also emphasized that students should critically evaluate AI-generated information instead of accepting it uncritically, reinforcing the teacher's continuing role in facilitating learning.
A Diversity of Questions and Creations
A Diversity of Questions and Creations
Participant Voices
Participant Voices
The reflections below illustrate key insights shared during the market discussion:
"As a postsecondary learning designer, I would choose NotebookLM when I need accurate, source-based support for research and course development. At the same time, Gemini is useful for brainstorming, generating ideas, and creating learning materials." — Manouchehr
"Notebook LM... has been really helpful for students that are auditory learners and has helped me with making connections across multiple sources and concepts." — Megan
"I think Claude did an excellent job summarizing the data into an easy to navigate mindmap on the left and a viewing pane for associated text on the right. It seemed to immediately understand the delivery method that would be naturally usable by readers." — Jake
Trade-offs Activity
Trade-offs Activity
Respondents were asked to give their design choices for creating an AILP with the following elements:
Help Level: How much of the thinking it does
Tone: Validate or push back
Initiative: Who drives the interaction
Memory & Personalization: Remember and adapt to each learner or not?
Summary of Themes
Summary of Themes
Productive struggle & learner independence: A major recurring theme is the desire to preserve "productive struggle." Respondents explicitly chose "Socratic" over "Hints" because they want to foster critical thinking, problem-solving, and independence rather than allowing learners to take shortcuts by getting direct answers.
Constructive feedback over blind praise: There is a strong consensus that feedback must be honest to be effective. Respondents noted that while support is needed to build confidence, too much praise without addressing weaknesses leads to a fear of failure and hinders actual skill development.
The privacy vs. personalization dilemma: The most conflicted choice was regarding memory. While almost everyone acknowledged that an AI is vastly superior when it can remember past interactions and personalize the learning pathway, this choice consistently brought up significant concerns regarding data privacy, surveillance, and the implications of permanent student records.
Context and classroom dynamics: Several respondents noted that the "best" choice is highly contextual ("it depends"). Designing for a single user is different than managing a classroom of 30 students with uniquely configured AIs. Factors like the learner's age, the specific subject matter (e.g., corporate training vs. grade school), and classroom routines heavily influence the ideal setup.
Overall, participants demonstrated that effective AILP design is less about maximizing AI capabilties and more about making intentional pedagogical decisions that balance support, independence, personalization, and learner agency.
Educational Value Activity
Educational Value Activity
After engaging with the research, participants revisited and refined their original AILP designs. The visual below highlights the common design shifts that emerged across the cohort.
Participant Voices
Participant Voices
The reflections below illustrate how participants refined their designs after engaging with the research.
"I shifted from an AI that primarily provided answers to one that uses questioning and conversation to encourage critical thinking, reflection, and independent problem-solving." — Jessy
"The design needs refining so that questioning and metacognitive prompts come before hints or direct support, preserving more opportunity for students to think independently." — Cherri-Lynn
"I refined my design to provide personalized guidance rather than direct answers because the research emphasized that AI should support learner agency rather than create dependence." — Jodee
"I shifted the AI's core programming away from immediate error-correction and instead implemented a Socratic questioning model. This ensures the tool focuses on long-term learning and critical thinking rather than just offering a temporary boost to immediate task performance." — Erika
Overall Reflection
Overall Reflection
While AILPs will continue to evolve, our cohort consistently returned to one central idea: the greatest educational value of AI lies in designing technologies that help learners think more deeply, reflect more intentionally, and maintain agency throughout the learning process.
Forecast Activity
Forecast Activity
Summary of Themes
Summary of Themes
Human-human connection remains essential: A strong consensus emerged that AI should enhance rather than replace teachers. Respondents across the board valued the uniquely human aspects of education and believed AI should free teachers to focus more on these areas.
Human-human connection remains essential: A strong consensus emerged that AI should enhance rather than replace teachers. Respondents across the board valued the uniquely human aspects of education and believed AI should free teachers to focus more on these areas.
Personalized tutoring is the most impactful pillar: Participants consistently envisioned AI becoming increasingly adaptive, providing individualized scaffolding and feedback while responding to learners' needs in real time. Many emphasized that AI should guide learners through questioning and productive struggle rather than simply providing answers.
Personalized tutoring is the most impactful pillar: Participants consistently envisioned AI becoming increasingly adaptive, providing individualized scaffolding and feedback while responding to learners' needs in real time. Many emphasized that AI should guide learners through questioning and productive struggle rather than simply providing answers.
Balancing innovation with ethical and social concerns: While participants were optimistic about AI's capabilities, some expressed concerns about privacy, data ownership, overreliance, and the potential erosion of authentic human interaction. Several expressed that various levels of governance and thoughtful implementation will be essential.
Balancing innovation with ethical and social concerns: While participants were optimistic about AI's capabilities, some expressed concerns about privacy, data ownership, overreliance, and the potential erosion of authentic human interaction. Several expressed that various levels of governance and thoughtful implementation will be essential.
Participant Voices
Participant Voices
The reflections below illustrate key insights shared during the forecast discussion:
"Rather than replacing peer interactions with personalized AI tutors, I see AI working behind the scenes to support educators and schools so they can create more meaningful human learning experiences." — Denise
"It may also match human-equivalent instruction... which to me is a bit concerning." — Gerta
"Personalized AI tutors will become more responsive by adapting in real time to how learners are performing, rather than simply following a predetermined pathway. For Pathfinder (my AI learning partner), this could mean analyzing a learner’s interview responses, identifying patterns in their communication or confidence, and dynamically adjusting practice scenarios and feedback based on their specific needs." — Jen
Page updated
Google Sites
Report abuse