Hi Everyone, I hope that this newsletter edition finds you well. As you will see from the below, we have been immensely privileged to host a number of events across the UK with some phenomenal partners. We have also welcomed our newest member of the GFF team, Mariia, and we have some insightful reflections to share from the wonderful Joseph Lin. As a reminder, everything that the Foundation offers is free to take part in and we are aiming to deliver our training and support schools all of the UK (and even beyond!), so if you have an interest in finding out what this could look like for your school community, don't hesitate to get in touch!
Welcoming GFF New Team Member!
We warmly welcome our new team member, Mariia to join us this month as our part time Marketing Designer! Mariia brings amazing talent in creating engaging visual content across digital media. We’re excited to see how her unique creative approach will enhance our organisation's visual presence!
Why Does AI Cost So Much? What Does the Money Get You?
In short:
Most people don’t have an accurate grasp of AI progress because free models lag far behind the state of the art.
Paid AIs can outperform free models by a wide margin (e.g., 43-point IQ gap), creating equity concerns in classrooms.
Contrary to older “tell-tale signs” of AI fakes, advanced systems have fewer flaws—and the first wave of AI literacy content may already be out of date.
The cost of AI is dropping fast, so while early adopters race ahead, others may still catch up in months rather than years.
I recently gave a school talk about information literacy in the age of AI, and we played with the Which Face is Real quiz from University of Washington’s Calling Bullshit project. The students were about 50% accurate when choosing AI-generated fakes, and the ones who did better than average were eager to point out telltale signs such as asymmetrical earrings, scrambled letters or logos, and other random defects.
Click on the image above to try the quiz yourself. It took me about a dozen tries to get the knack of it. Source: whichfaceisreal.com
But there were also those who got all of them wrong, who struggled to identify AI fakes even after learning about the usual tells.
It gets worse: the images in the quiz were generated with the seven-year-oldStyleGAN model, which is ancient by modern AI standards. As of 2025, it is trivial for models to generate pictures indistinguishable from reality, although many platforms block that capability with safety filters. However, intensifying competition and complaints from users around censorship have driven model providers to relax those guardrails, as OpenAI and xAI have done in the past month. Nowadays, it’s easy to generate AI images that easily pass as real, especially if you’re just scrolling past them on social media:
Realistic AI Fake: Amateur composition and blurry smartphone styles are common realism tricks. Source: Reddit
The Good Future Foundation has a paper in the works on AI and misinformation, so I defer to my colleagues to explore that aspect. Instead, I’d like to highlight that this phenomenon of paid AIs being more powerful than free ones is even more true for text generation, which means well-to-do students can get a significantly more advanced tutor, which is also harder for AI detectors to flag. (That said, AI detectors don’t work, and many experts recommend banning them.)
Case in point, consider the staggering 43-point IQgap between GPT-4o (the default model on ChatGPT’s free plan) and o1 Pro (which is exclusive to ChatGPT Pro users paying 200USD/month):
Yes, IQ tests can be problematic and some may disagree with “humanising” them by comparing them against human intelligence in this way, but the comparative point stands. Source: trackingai.org
This paywall on intelligence is a problem not just because of inequity; because free AIs are often a few steps behind cutting-edge models, most people don’t have an accurate grasp of just how powerful and disruptive AI models have already become:
OpenAI’s expensive o1 model got a “perfect score” on a Carnegie Mellon undergraduate math exam. The professor who designed the problems noted that this cost “around 25 cents, for work that most people can't complete in 1 hour.”
Turns out, you do get what you pay for when it comes to AI, and we are nowhere near the point of diminishing returns because there are now two clear paths for AI to become smarter (by training larger models, and by letting models reason for longer).
I’m an optimist, and I think AI tutors will solve Bloom’s two sigma problem: the cost of AI has been falling 10x every 12 months, so while early adopters can definitely race ahead, most others can still catch up in months rather than years.
AI Coding: No Coding Required!
In Short:
Design Thinking Meets AI: Students follow a standard design-thinking approach, but AI handles many routine tasks.
Lower Barrier to Entry: Secondary students use generative AI tools (e.g., Replit Agent, AI search engines) to prototype real sustainability solutions.
Focus on Agency & Integrity: Research suggests “cheating culture” and motivation matter more than AI usage frequency in predicting plagiarism.
One of my favorite recent studies examined ChatGPT and plagiarism in Spain. The authors found that “a higher frequency of using ChatGPT for academic purposes does not necessarily lead to a higher likelihood of plagiarism,” instead it is “certain personal variables…such as being involved in a cheating culture or experiencing amotivation, that cause the likelihood of plagiarism to increase.”
I love this conclusion because it means teachers have a powerful role to play, to influence students to use their newfound AI empowerment for good. The challenge lies in creating a school culture that promotes agency, integrity, and empathy; furthermore, as more and more of knowledge acquisition is done through AI tutors, there is a strong case to be made that enculturation will become the most important contribution of education to society.
A very direct way to do so is to teach students to build something that supports Sustainable Development Goals. This has been a typical assignment at the undergraduate level, but GenAI assistance has made it feasible for secondary students to try their hands at similar problem-based and project-based learning. Here’s a course I’m currently piloting, which uses AI to help non-coders create local SDG solutions.
Overview
Course Theme: AI Coding x Sustainability Solutions
Grade: Form 3-5
Teacher/TA/student ratio: 1 vs. 3 vs. 30
Duration: 24 hours
How It Went Down
This course is essentially a design thinking course, except everything is sped up because a lot of the work is done by AI. Nevertheless, students have to understand and synthesise information in order to deliver their presentation: the tasks are complex enough and spread across several AIs, that they cannot just prompt the AI once and hand in something AI-generated.
What Students Built
The course’s learning focus isn’t really the coding, rather it is the skill or mindset of using AI to study a problem, think of solutions, then turning them into reality. The primary assessment criteria is in how well they explain and justify their ideas, followed by how well their solutions work.
Here are some ideas that the students had:
SDG 3: Ensure healthy lives and promote well-being for all at all ages
An AI menu analyzer to help people choose low-sodium meals
A free cognitive test for the elderly based on peer-reviewed content, to detect cognitive decline
SDG 4: Ensure inclusive and equitable quality education and promote lifelong learning opportunities for all
An app for students with reading disorders, to hear their textbooks read out loud by taking a picture of the page and voicing it with AI
What I Learned
It was a great experience. Teaching design thinking and thematic studies this way is almost relaxing because AIs take care of a lot of the students’ simple questions. Other than the usual sessions introducing concepts and explaining how to use the AI tools, much of the class hours were spent on self-directed research, building, and presentation preparation.
My TAs and I rotated through the groups, made sure they were on task, gave advice and deeper explanations when needed, and relied on regular presentation rehearsals every other lesson to establish cadence and momentum.
Constant prodding from trainers plus the “threat” of having to stand up to present to class was enough to keep the students working, and I’d rate most of their output at a weak-to-mediocre undergraduate level—for high schoolers, that’s already an impressive achievement.
If I had to scale this to larger classes or fewer teachers, I would reduce the open-endedness, such as preparing a set of ideas/apps they have to choose from instead of letting them freestyle; that allows me to curate relevant tutorial videos for them to follow, essentially using YouTube to replace TAs.
In the next edition…
Teaching students how to prepare for interviews and polish their self-introductions with AI.
Training and Dialogue to Move AI in Education Forward
Professional Development on the Road and BBC Coverage
March has been an energetic and rewarding month for GFF as we brought our free-to-attend AI professional development sessions to Croydon, Stoke-on-Trent, and Rochdale alongside our partners at Educate Ventures. We've been inspired by teachers' enthusiasm and dedication, many of whom are deeply committed to understanding AI's role in teaching and learning in order to equip their students with the knowledge and skills needed to navigated an AI-infused world with confidence. Our sessions which put AI literacy, compliance and data privacy high on the agenda have also attracted wider community interest, including BBC News coverage which shows the growing public awareness of AI's significance in education.
A heartfelt thank you to our local partners — UTCAI and #MerkyFC HQ in Croydon, Belgrave St. Bartholomew's Academy in Stoke-on-Trent, and the Watergrove Trust in Rochdale. Their amazing support has played a key role in starting the conversation about AI education in their communities!
The collaborative spirit was truly electric as educators gathered to explore the future of learning. Partnering with good future foundation and Educate Ventures Research we dove deep into how artificial intelligence—especially natural language processing—is revolutionising education. I’m sure this conversation is just beginning.
- Troy Davis, Co-Founder of UTCAI
The Ai event hosted by Good Future Foundation shared the amazing potential of AI if we harness it to empower children. We can’t predict the jobs of the future but we can equip children with the skills needed to be successful. Thank you for an amazing event attended by leaders from across the city.
- George Barlow, Principal of Belgrave St. Bartholomew's Academy
Generative AI can be a double edged sword, and particularly in education we need to appreciate both the challenges and the benefits that it can bring. At Watergrove Trust we are cautiously optimistic about how it can be used to improve outcomes for learners and reduce staff workload, so the opportunity to work with Good Future Foundation and Educate Ventures Research was not to be missed. Over 120 colleagues attended the sessions and feedback has been overwhelmingly positive. The day has energised our work in this field and equipped us to make better use of the tools available.
- David Leonard, Strategic IT Director, Watergrove Trust
Next Stop at Queen Ethelburga's Collegiate in York
Just like Joseph mentioned in the earlier article, AI has been developing at lightning speed and creating both powerful opportunities and disruptions. We need collaborative efforts within and across sectors to navigate these changes and thoughtfully address both the benefits and risks that this fast evolving technology brings.
GFF has been privileged to participate in many important conversations with like-minded organisations focussing on AI in education in the past few weeks. Events we participated in include:
The Society of Heads Annual Conference in which we encourage independent schools to collaborate in making AI-focussed professional development more equitable for all teachers.
The Chartered College of Teaching Annual Lecture where a diverse line up of educators discussed evolving best practices as AI drives us to reflect on the meaning of education.
“With developments of AI leading to new opportunities and risks for schools at an almost unprecedented rate, it was vitally important to host a conversation that allowed us to go deeper into discussions around ethics, strategy, the digital divide, rather than focusing on tools and tips. The positive feedback from attendees, as well as the panel members, has shown how valuable they found the chance to explore these issues and connect with others around AI in schools. We are hugely grateful to our sponsors, The Good Future Foundation, and our hosts, Salesforce, for their support with the event.”
- Dr Cat Scutt MBE, Deputy Chief Executive of Chartered College of Teaching
The Research and Evidence Forum at Eton College where we shared AI’s impact on social emotional learning and heard from other presenters on other informed research around the state of play concerning AI in education.
The Research and Evidence Forum that Eton’s Centre for Innovation and Research in Learning (CIRL) hosts provides an opportunity for educators to come together and hear each other present their evidence-based reflections on developments in teaching and learning. The most recent forum focused on AI in education: how educators are using it, what are the ethical, evaluative and compliance issues for implementing it in the classroom and across the whole school, what social and emotional impact might it have, and what are the opportunities it offers. We were delighted that Daniel Emmerson joined us to talk about the values-driven work that the Good Future Foundation does in this area to help schools to prepare their pupils to succeed in an AI-enhanced future. Their work is particularly important because the opportunities provided by AI risk making the digital divide ever wider, and the Foundation enables pupils no matter what background they are from.
- Jonnie Noakes M.A., Director of Teaching & Learning, Eton College
The future of AI holds many unknowns. No single organisation has all the answers about how it will develop and unfold. Therefore, we must explore this landscape together, learning from each other’s perspectives and experiences as we collectively contribute to this expansive field.
AI Quality Mark Highlights
Award-Winning School Spotlight
If you're still considering whether to pursue the AI Quality Mark, take a moment to check out this testimonial from King’s School Rochester who has recently achieved the Bronze Award. It might perhaps inspire you to take this exciting step.
The Good Future Foundation’s AI Quality Mark process has been an invaluable experience for our school. Achieving the Bronze award provided us with a clear and structured framework to assess our current AI integration and identify areas for growth. The process was well-organized and encouraged meaningful reflection on how AI can enhance teaching, learning, and operational efficiency. The structured self-assessment and external moderation gave us confidence that we are on the right track while also helping us pinpoint specific strategies for improvement.
Beyond the accreditation itself, the Quality Mark has given us a solid benchmark to guide our future AI initiatives. The insights gained through the assessment have informed our strategic planning and policy development, ensuring that AI is embedded in a way that is both responsible and effective. We now have a clear roadmap for expanding our use of AI across the curriculum and operational aspects of the school, and we look forward to continuing this journey with the support of the Good Future Foundation.
-Coetzee CZC, the Director of Digital Strategy, King’s School Rochester
We can’t stress enough that the unique self-evaluation combined with constructive moderation support is just as valuable as the award itself. Many schools are just starting out and are beginning with the Progress Award as they develop their own AI strategies that align with their priorities. We look forward to supporting more and more schools along the way.
A Community Where Schools Share Best Practices
We recently launched our first bi-monthly online community meeting for schools that are going through the AI Quality Mark where teachers could share, ask questions, and learn how other schools approach issues around AI. In this first meeting, we are incredible grateful to have the team from Brentwood School to genuinely share with other fellows their journey so far.
Students-led Dialogues on Issues Around AI
Our brilliant Student Council just kicked off the first session of the Online Global Student Dialogues earlier this month, and engaged with students from various countries to discuss how AI is shaping the world and their futures.
The second session of the Dialogues is set for 24th April. Student participants will discuss how to balance using technology for help with intellectual growth, looking at our reliance on AI, its effects on how we learn, and the possible dangers of misinformation. We already have students from England, the US, and Ghana signed up for the event. If you're also interested in having your students join a conversation about responsible AI use in schools, be sure to register!
Student Council met to learn the facilitation skills with Melissa from GEBG to prepare the Student Dialgoues
Foundational Impact Podcast: Sharing from Suzy Madigan and Student Council
This month’s podcast amplifies voices that are often overlooked in the AI narrative. We hear from Suzy Madigan, Responsible AI Lead at CARE International, who shares insights from her report on AI's impact in the Global South. In the other episode, four members of our Student Council sharing their firsthand experiences and observations about generative AI use among young people.
Both of these episodes provide great insights into AI's impact across different communities and generations. Listen now to explore both the global implications of AI and its influence on today's youth!
