AI for Educators Daily with Dan Fitzpatrick

Dan Fitzpatrick, The AI Educator

Hey, I'm Dan, The AI Educator. I know that we both care deeply about the state of education, amid the uncertainty of rapidly advancing AI. I work with leading schools and governments worldwide to help them strategise and build capability, and I have recently been recognised as a top voice on AI. While most teachers are aware of the influence of AI on education and student learning, many are unsure how to respond in practice. My mission is to amplify credible expert insight and give educators the clarity, confidence, and tools they need to teach effectively and prepare students.

  1. Maximizing capability gains for AI agents

    21 hr ago

    Maximizing capability gains for AI agents

    Send us Fan Mail An AI system built software in 14 hours for $251, a task normally taking a human team months. This is AI agents in education. In this episode: An AI system, Opus 4.7, accomplished a complex software development task in 14 hours that typically requires human teams months, showcasing dramatic AI capability gains.The way we interact with AI is evolving from co-intelligence chatbots to autonomous AI agents that complete complex tasks with minimal human oversight, shifting the focus to managing AI in schools.Ethan Mollick emphasizes that domain expertise is crucial for maximizing the effectiveness of AI agents, as experts achieve better results when assigning AI for complex tasks.Educators must adapt their approach to AI literacy, moving beyond basic prompting to teaching students how to critically manage AI agents, thereby preparing them for the future of AI work.The exponential growth of AI means schools must implement continuous professional development for teachers to strategically leverage AI for complex tasks and navigate constant technological shifts.Chapters: 00:00 — Cold open & welcome00:30 — Opus 4.7: Staggering AI capability gains01:00 — Measuring AI's ability to do real work01:45 — Near-frontier AI models and their exponential growth02:30 — Implications for AI agents for educators and workflows03:15 — From chatbots to AI agents: A changing interaction model04:30 — Managing AI in schools: The manager's role in the future of AI work05:45 — Rethinking AI literacy and collaborative reasoning06:45 — The impact of exponential AI capability gains on institutions07:30 — Designing professional development for continuous adaptationWhat are AI agents and how do they differ from chatbots for educators? AI agents are long-running, smart, self-correcting AI systems that tackle complex tasks with less human intervention than traditional chatbots, requiring educators to shift from interactive prompting to managing AI workflows. How can teachers use AI agents for complex tasks in the classroom? Teachers can assign AI agents to handle time-consuming administrative tasks, content generation, and aspects of differentiation, freeing up human capacity for unique human skills like judgment and relationship-building. What is the future of AI work for students and how should schools prepare them? The future of AI work involves managing AI agents rather than just using chatbots, so schools should teach students to critically evaluate AI outputs, identify biases, and apply human judgment to AI-generated insights. Featuring: Dan Fitzpatrick, Ethan Mollick, Claude Fable, GPT-5.6, METR, UK’s official government AI Security Institute, GDPval, Epoch, Opus 4.7. Read the original source Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    11 min
  2. Only 2% have an AI strategy

    1 day ago

    Only 2% have an AI strategy

    Send us Fan Mail Only 2% of secondary schools in England have a comprehensive AI strategy, despite AI already being embedded in teaching and learning. In this episode: A new report reveals only 2% of secondary schools in England have a comprehensive AI strategy, despite widespread informal adoption for tasks like lesson planning.The primary barrier to implementing AI strategy schools England is a lack of staff confidence and skills (63%), not financial cost.School leaders who actively engage with and model AI use foster more consistent AI adoption across their institutions and are key to effective AI staff training schools.A clear AI policy education should focus on 'purpose over technology,' defining *why* AI is used, not just *how not to*.Uneven AI integration and AI staff training schools, particularly between London and other regions, risk deepening educational disparities in England.Chapters: 00:00 — Cold open & welcome00:25 — Only 2% of schools in England have an AI strategy00:55 — Distinguishing AI policy from AI strategy in education01:30 — Current AI uses for lesson planning and efficiency02:00 — Leadership engagement with AI and perceived risks02:40 — Biggest barriers: staff confidence and AI literacy03:15 — Impact of leadership modelling on AI adoption03:55 — Regional disparities in AI use and widening inequality risks04:50 — Five practical steps for an AI strategy schools England06:50 — Seizing opportunity through purposeful AI strategyHow many schools in England have a comprehensive AI strategy? Only 2% of secondary schools in England have developed a comprehensive AI strategy, according to a report by Accenture and Teach First, despite AI already being embedded in teaching and learning. What are the biggest challenges for schools implementing AI? The primary challenge for schools implementing AI is a lack of staff confidence or skills (63%), followed by data privacy concerns (51%) and a limited understanding of AI's educational potential. How can school leaders encourage AI adoption among staff? School leaders can encourage AI adoption by directly engaging with AI, demonstrating responsible use, allowing controlled experimentation, and fostering ongoing shared learning among staff. Featuring: Dan Fitzpatrick, Accenture, Teach First, Matt Prebble, James Toop, England, London. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    10 min
  3. Fable 5's return

    2 days ago

    Fable 5's return

    Send us Fan Mail The most capable AI model ever released was shut down by the US government then returned, a wild saga with vital lessons for AI policy for schools. In this episode: Anthropic's Claude Fable 5, the "most capable AI model ever released," experienced a rapid launch, government-mandated shutdown, and return, offering a blueprint for future AI model reliability challenges.The 22-day saga of Claude Fable 5 underscores that a robust AI policy for schools must include contingency planning to avoid a single point of failure when integrating AI into workflows.The US government's swift intervention with emergency export controls on Fable 5 signals that government AI regulation is a dynamic landscape, necessitating adaptable AI in education strategy based on principles, not specific tools.Fable 5's comeback, facilitated by enhanced safety classifiers and a HackerOne program, illustrates that transparent, iterative improvement in AI safety is achievable and builds trust.Schools should model Anthropic's transparency by communicating openly about AI tool limitations and actively working to improve them, fostering reflective awareness of AI's capabilities and constraints.Chapters: 00:00 — Cold open & welcome00:30 — The wild saga of Claude Fable 5's launch and shutdown01:25 — Understanding Anthropic's Mythos-class models and safeguards02:40 — Staggering early performance and pricing of Fable 503:20 — Why the US government pulled the plug on Fable 504:35 — The engineering and diplomacy behind Fable 5's return05:40 — Lesson 1: Contingency in AI policy for schools – avoiding single points of failure07:15 — Lesson 2: Adapting to government AI regulation with principle-based policies08:25 — Lesson 3: The hopeful truth about AI model reliability and transparency09:40 — The enduring task for educators in an age of extraordinary AI capabilityWhat is Anthropic Fable 5 and why was it temporarily shut down? Anthropic Fable 5 was a highly capable AI model launched by Anthropic that was temporarily suspended by the US government due to national security concerns after researchers bypassed its safeguards. How does the Fable 5 incident inform AI policy for schools? The Fable 5 incident teaches schools to build contingency plans for AI tools, assume dynamic government AI regulation, and anchor their AI in education strategy in adaptable principles rather than specific products. What does the Fable 5 story tell us about AI model reliability and safety? Fable 5's return demonstrates that AI model reliability can be enhanced through dedicated engineering work on safeguards, transparent collaboration with regulators, and open communication about limitations, proving safety and capability are not mutually exclusive. Featuring: Dan Fitzpatrick, Anthropic, Claude Fable 5, Project Glasswing, Claude Opus 4.8, Mythos 5, US government, HackerOne, OpenAI GPT-5.5. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    14 min
  4. DeepMind’s surprising Sierra Leone trial

    3 days ago

    DeepMind’s surprising Sierra Leone trial

    Send us Fan Mail An AI tutor in Sierra Leone reportedly helped students gain a year of schooling in eight weeks, a claim even Google DeepMind cautions us to take with a grain of salt. In this episode: Google DeepMind's AI tutor education trial in Sierra Leone estimated students gained a year's learning in eight weeks, using a re-engineered Gemini model.The AI learning tool, Guided Learning, was designed not to give direct answers, fostering productive struggle crucial for learning outcomes.Early findings suggest AI in classrooms might initially widen the gap for already proficient students, prompting Google DeepMind to investigate different pedagogies to support struggling learners.Teachers in the Sierra Leone trial adapted quickly to AI learning tools, discovering new teaching strategies and increasing one-on-one student interaction after just one day of training.Transparent research practices from Google DeepMind, including a public playbook and training materials, aim to allow wider adoption and scrutiny of AI for learning outcomes.Chapters: 00:00 — Cold open & welcome00:45 — Irina Jurenka's shift to social impact at Google DeepMind01:45 — The core tension: AI as assistant vs. AI for learning02:45 — How Guided Learning became a true AI tutor education tool03:45 — Overcoming the AI's tendency to give answers04:45 — The 'take with a grain of salt' claim on learning gains05:45 — Addressing equity and the 'widening gap' concern in AI for learning outcomes06:45 — Google DeepMind's transparency and research integrity07:45 — Transforming teaching practices with AI learning tools08:45 — The future evolution of AI in classroomsHow did Google DeepMind's AI tutor impact learning in Sierra Leone? The AI tutor education trial in Sierra Leone estimated students gained approximately one year of schooling in just eight weeks, though this figure comes with a scientific caveat from the researchers. Can teachers trust AI learning tools not to give students answers directly? The Guided Learning tool, designed by Google DeepMind for AI in classrooms, was specifically engineered not to give direct answers, forcing students into productive struggle, unlike many public AI models. Do AI learning tools widen the learning gap, or help all students? Initial findings from the Sierra Leone trial showed that stronger math students benefited most, but Google DeepMind is actively researching different pedagogical approaches for AI to ensure it helps raise the floor for struggling students too. Featuring: Dan Fitzpatrick, Irina Jurenka, Google DeepMind, Sierra Leone, Guided Learning, Gemini, World Bank, Stanford, Nigeria. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    14 min
  5. Navigating AI-generated content

    6 days ago

    Navigating AI-generated content

    Send us Fan Mail 70% of people can't spot AI deepfakes, leaving students vulnerable to hidden AI in advertising without proper AI literacy. In this episode: A *Guardian* investigation revealed that 70% of people cannot detect *AI deepfakes*, making students vulnerable to hidden *AI in advertising*.Brands like *Once* and *Maket* are using undisclosed *AI generated content* and *AI-generated influencers* due to lower costs and fewer risks compared to human talent.Current *AI transparency rules* are lagging; the UK's *Advertising Standards Authority* doesn't explicitly prohibit undisclosed AI, although the EU will require labeling.Developing *AI influencers ethics* and critical AI literacy is paramount for educators to equip students to navigate a world saturated with AI-generated information.Educators must foster critical thinking skills, including basic *AI deepfake detection*, and discussions around *AI transparency rules* in the classroom.Chapters: 00:00 — Cold open & welcome00:30 — The Guardian's investigation into undisclosed AI in advertising01:15 — Case studies: Once, Maket, and Ashle using AI-generated influencers02:15 — Educational implications: Why AI literacy is crucial for students03:00 — Why brands use AI: Cost savings and managing public image03:45 — The blurring lines of authenticity and 'plausible deniability' in AI content04:30 — Regulatory response: Which?, Lisa Barber, and the Advertising Standards Authority05:15 — The urgent need for AI deepfake detection and AI transparency rules in education06:00 — Cultivating human judgment and ethics over technological prowessHow can teachers equip students to identify *AI deepfakes* and *AI generated content*? Teachers can equip students by fostering AI literacy as a critical thinking skill, teaching them to question AI outputs, biases, and intent, and discussing *AI transparency rules* in the classroom. What are the ethical concerns surrounding *AI in advertising* and *AI influencers ethics*? Ethical concerns include misleading consumers with undisclosed *AI generated content*, eroding trust, and the potential for manipulation when 70% of people cannot detect fake videos. Are there *AI transparency rules* or regulations for brands using *AI-generated influencers*? Currently, the UK's *Advertising Standards Authority* does not explicitly prohibit undisclosed *AI generated content*, though the EU's new Artificial Intelligence Act will require deepfakes to be labeled. Featuring: Dan Fitzpatrick, Once, Maket, Ashle, Reality Defenders, Get Real Labs, Which?, Advertising Standards Authority, Lisa Barber. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    8 min
  6. AI Critical Thinking Education: Addressing Bias in Classroom AI

    2 Jul

    AI Critical Thinking Education: Addressing Bias in Classroom AI

    Send us Fan Mail Almost 30% of Saudi teachers already correct biased AI outputs, highlighting an urgent need for students to interrogate AI, not just trust it. In this episode: A 2025 Saudi Arabia survey found nearly 30% of teachers are already correcting AI bias in education, highlighting an urgent need for students to interrogate AI.The core issue: most AI tools are trained on English-language and Western-dominant datasets, creating linguistic and cultural blind spots in AI-generated knowledge for diverse learners.Teachers must evolve into 'epistemic intermediaries,' guiding students in AI critical thinking education by modeling how to assess AI outputs for accuracy and cultural relevance.True AI literacy for students involves collaborative reasoning and actively critiquing AI responses, not just passively accepting them.Designing assessment tasks around "Product, Process, and Performance" can ensure students engage in cognitive stretch, applying unique context and judgment, which cannot be faked by AI tools.Chapters: 00:00 — Cold open & welcome00:25 — Saudi Arabia's AI bias challenge: 30% of teachers correcting AI00:55 — Cultural and linguistic blind spots in AI tools01:50 — AI critical thinking education: shifting from teacher as authority to AI interrogator02:45 — The teacher's new role: 'epistemic intermediary' assessing AI outputs03:50 — Redefining AI literacy for students: collaborative reasoning and critique04:45 — Saudi Arabia's proactive approach to addressing AI bias in education05:25 — School leaders: prioritizing AI critical thinking education over technology adoption06:10 — Protecting human judgment, imagination, and wisdom in responsible AI in education06:45 — Knowledge transmission to knowledge interrogation: The core shiftHow can teachers address AI bias in education in their classrooms? Teachers can address AI bias by becoming 'epistemic intermediaries,' systematically assessing AI-generated content with students for factual accuracy, linguistic precision, cultural relevance, and contextual appropriateness. What does AI critical thinking education look like for students? AI critical thinking education involves teaching students to systematically critique AI responses, compare outputs across languages, identify inconsistencies, and consciously inject missing cultural nuance into AI-generated content. Why is responsible AI in education crucial for non-English dominant contexts? Responsible AI in education is crucial because most AI tools are trained on English-language and Western-dominant datasets, leading to inherent linguistic and cultural blind spots that can misrepresent local realities for students in other regions. Featuring: Dan Fitzpatrick, Basmah AlBuhairan, Reem Taibah, Amani AlOlayani, King Abdulaziz City for Science and Technology, Centre for the Fourth Industrial Revolution Saudi Arabia, Ministry of Education, Saudi Arabia, World Economic Forum. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    9 min
  7. AI and oracy skills: The most valuable graduate skill in the AI age

    1 Jul

    AI and oracy skills: The most valuable graduate skill in the AI age

    Send us Fan Mail 94% of students used AI in assessed work by 2026; learn why AI weakens core knowledge vital for genuine oracy. In this episode: A 2026 Higher Education Policy Institute survey revealed 94% of students used AI in assessed work, necessitating a renewed focus on oral communication education.Bruce Hood argues that AI, while not directly harming oracy, weakens the core knowledge and synthesis abilities vital for genuine oral communication, impacting graduate employability skills.Employers now prioritize verbal communication and soft skills over grades, highlighting the increasing importance of assessing oral communication for future job markets.The Three Minute Thesis (3MT) competition, pioneered by the University of Queensland, demonstrates a successful model for developing advanced oral communication skills in students.Educators should shift from solely written assignments to incorporate more oral assessments to build essential AI and oracy skills from primary school through higher education.Chapters: 00:00 — Cold open & welcome00:25 — 94% of students using AI in assessed work00:55 — Employer demand for oral communication education01:30 — How AI weakens core knowledge for AI and oracy skills02:25 — The strategic choice for integrating oral communication education03:05 — Designing learning tasks for oral communication beyond written reports03:50 — Assessing oral communication: Product, Process, and Performance04:30 — Embedding oracy skills through professional development and AI Mavericks05:15 — The Three Minute Thesis (3MT) as a model for developing graduate employability skills05:55 — Oracy as the human differentiator in an AI-augmented worldHow does AI impact student skills for oral communication education? AI can weaken the foundational knowledge and deep synthesis required for effective oral communication by performing too much of the cognitive heavy lifting that students previously did themselves. What graduate employability skills are most valued by employers in the AI age? Employers increasingly value verbal communication, critical thinking, authenticity, and the ability to establish credibility and respond on the fly, as reported by the National Association of Colleges and Employers and CBI Economics. How can educators improve assessing oral communication in the AI era? Educators can shift focus to oral assessments like debates, presentations, conversational exams, and performance-based tasks that require students to apply knowledge dynamically and respond to spontaneous questions. Featuring: Dan Fitzpatrick, Bruce Hood, Higher Education Policy Institute, National Association of Colleges and Employers, CBI Economics, University of Queensland, Three Minute Thesis, My-Thesis, University of Bristol. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    9 min
  8. AI for underserved classrooms: Powerful learning with no internet

    30 Jun

    AI for underserved classrooms: Powerful learning with no internet

    Send us Fan Mail An AI maths tutor on WhatsApp boosts learning a year for $5/child, proving AI for underserved classrooms doesn't need fast internet or expensive tech. In this episode: An AI maths tutor named Rori on WhatsApp is achieving a year of learning gains for just $5 per child, demonstrating powerful AI for underserved classrooms without needing fast internet.Solutions like FoondaMate and Juza AI prove that offline AI education and low-resource AI learning can effectively support millions of students in developing countries.A 2025 UNESCO paper advocates for an 'offline-first' principle for scaling AI in African schools, highlighting the importance of resilient edtech in developing countries.Digital Public Infrastructure (DPI) can provide shared resources like local-language corpora and governance rails, enabling more effective and equitable AI for underserved classrooms to scale.Pedagogy and teacher trust are non-negotiable; AI tools must align with how children learn and enhance, not replace, the teacher's role for successful implementation.Chapters: 00:00 — Cold open & welcome00:30 — Rori: AI for underserved classrooms on WhatsApp01:30 — Challenging assumptions: AI doesn't need perfect infrastructure02:45 — The paradigm shift to low-resource AI learning03:45 — Evidence of AI in African schools: FoondaMate, GlobeDock, Juza AI04:45 — Building for constraints: Offline AI education and edge computing benefits05:45 — The missing scaffolding: Education Digital Public Infrastructure (DPI)07:15 — Non-negotiables: Pedagogy and the teacher's role08:30 — Strategic investment for AI in African schools09:00 — Conclusion: AI for the world as it isHow can AI support learning in underserved classrooms without internet? AI tools like Rori, FoondaMate, and Juza AI demonstrate that effective learning can occur using simple phones, minimal data, or entirely offline 'AI in a box' solutions by syncing data periodically or running models locally. What are the key benefits of low-resource AI learning for students? Low-resource AI learning provides accessible, affordable, and high-quality educational support to a broad range of students, particularly the 'middle 80%' who may not have access to traditional tutors or high-tech solutions. What is Digital Public Infrastructure (DPI) and how does it help scale edtech in developing countries? DPI creates a shared, public layer of educational resources such as curriculum-aligned content, local-language data, and governance standards, allowing local developers to build scalable, safe, and interoperable AI solutions. Featuring: Dan Fitzpatrick, WhatsApp, Rori, FoondaMate, World Bank, Messenger, GlobeDock Academy, Juza AI, UNESCO. Follow AI in Education with Dan Fitzpatrick for more on AI in education.

    11 min
See All (325)

About

Hey, I'm Dan, The AI Educator. I know that we both care deeply about the state of education, amid the uncertainty of rapidly advancing AI. I work with leading schools and governments worldwide to help them strategise and build capability, and I have recently been recognised as a top voice on AI. While most teachers are aware of the influence of AI on education and student learning, many are unsure how to respond in practice. My mission is to amplify credible expert insight and give educators the clarity, confidence, and tools they need to teach effectively and prepare students.

Information

You Might Also Like