Today, we'll be asking whether advances in artificial intelligence are a threat to teachers. Are the machines coming for our jobs? The AI Academy podcast is a weekly podcast helping educators like you leverage AI In your everyday practice, I'm your host, Andy Fisher, and thanks for joining me. The question of whether artificial intelligence will replace teachers in the classroom is, of course, an emotive one.

However, I'd like to invite you for the next 20 minutes or so to look at this as an opportunity to reflect on what it is that we do, whether we are replaceable. Where we add value and how we might be able to leverage emerging technology so that it remains our servant rather than our master. In other words, we need to ask the question, even if artificial intelligence could replace teachers, should it? Is that what we want for our future? 

Now I came across a charming quotation by Joanna Maciejewski, who states, "I want AI to do my laundry and dishes so that I can do art and writing, not for AI to do my art and writing so that I can do the laundry and dishes." And I think that's absolutely spot on. It's so easy to lose sight of why we are deploying technology, what we want it to do for us.

I'm going to be drawing. Fairly significantly on a LinkedIn article that I read recently by Chris Goodall, who's the head of digital education at AI in Education. And I'll put a link in the show notes to the original piece and in it, Chris identifies four questions that need to be considered when addressing the degree to which AI is a threat to teachers. 

The first is that we need to clearly define what is the role of a teacher. What is it that we do? Then we should ask how rapidly is artificial intelligence developing? In other words, how quickly might we be at a point when AI could replace us? Third, we have to question the quality of learning and the level of human connection that we can currently provide to students. Is the current teaching model fit for purpose? And then finally, We need to reflect on the quality and supply of teachers in the profession right now and in the foreseeable future. 

So in terms of what we do, obviously, I'm not going to patronize you here. We know how we spend our days, evenings, weekends, but the role of a teacher is also very clearly articulated in the UK teaching standards. There are eight of them, and I think it would be useful for this conversation if we just remind ourselves of what they are. 

So standard number one says that teachers set high expectations and challenge for our pupils to that. Two, we should ensure that students progress and that they are guided to work towards specific outcomes. The third is that teachers should demonstrate excellent subject knowledge. Fourth, we should be able to deliver well structured lessons. Fifth, we should be able to adapt those lessons to the specific needs of the students in the classroom. Six, we should have effective assessment methods. Seven, we should be able to provide a safe environment and manage the behavior of the pupils in our class and finally, eight, we should be involved in the broader, holistic demands of education, including reflecting the ethos of our school, In the way that we interact with pupils and through professional collaboration with colleagues.

Now, many of those involved in the cutting edge of EdTech recognize that adaptive educational platforms using AI, and we'll come back to those later, can already meet the demands of the first six teaching standards. But it's standards seven and eight where Chris Goodall argues, and I agree with him, that at least for the time being, Artificial intelligence is not able to match or surpass the level of provision offered by human teachers.

And this distinction is going to become the lodestone for the rest of what I have to say today. I don't think that teachers can or will be replaced by artificial intelligence, at least in the short to medium term, because as we know all too well, The transference and testing of knowledge is a small part of what we do.

It's not why most of us came into the profession or remain in it despite the challenges it presents, a good teacher is a mentor and a coach. We inspire, we spark conversations. We model what it is to be hopefully a mature and autonomous, sovereign being, and not just a regurgitator of information. And I'll return to this at the end of the episode.

So let's address the question of how rapidly AI is developing. And to do that, we need to have a better understanding of exponential growth and how long it will take before AI can match not one, but all of the outcomes achieved by teachers in terms of their academic performance. 

Alan Turing published his paper, 'Computing Machinery and Intelligence', just 75 years ago, and he had a visionary perspective, but I think even he would be shocked at how quickly AI systems have evolved.

And 15 years later, Gordon Moore, the co founder of Intel, made his prediction, which has come on to become known as Moore's Law. It states that the number of components that can fit on a single computer chip can be expected to double every two years, making computers increasingly small and faster.

And this has held true for the last 70 years, until now we have engineers producing components that are size capped because they are literally working at the atomic level. And as computers have become more powerful, the software and systems that they can run have shared a parallel acceleration in terms of performance, including AI systems.

In standardized testing AI models can now match or exceed PhD qualified humans in visual reasoning, reading comprehension, competition level mathematics, STEM subjects and coding. And this progress shows no sign of slowing down. So assuming that innovation isn't stymied by state regulation or black swan events, we can expect AI systems to exceed all human abilities in domains of knowledge in the near future.

So with that in mind, let's now consider the degree to which our current educational models are fit for purpose. And it's this that gives me the most pause to reflect on my own teaching practice. We know that the concept of a modern school was originally created to educate the masses as part of an empire building campaign in the Industrial Revolution.

And it did so very successfully, creating the administrators and the managers of the future generations, whilst also providing childcare so that eventually both parents could enter the workforce. Our schools are inextricably bound to political and economic policies, and pedagogy has always had to manage this somewhat strained relationship.

A system like this, with its economies of scale, has many advantages. A broad and liberal education is no longer the preserve of the wealthy aristocracy. It's free and it's founded on equity and inclusivity, but it does come with its limitations. 

Having worked alongside teachers in the maintained and independent sectors in the UK, I can see the pinch points and the pressures that are mounting on us to be able to effectively deliver a curriculum where class sizes, time constraints and budgets make it nearly impossible to achieve the set targets.

In fact, Chris Goodall quips in his article that you'd need to be a machine to cope with the demands of what it means to be a teacher in the 21st century. We see this sadly in the numbers of educators with burnout, those suffering poor mental and physical health, and in the recruitment and retention challenges that plague our profession.

It's not an easy job. And most would argue it's just getting more challenging. Now, I want to be really clear here. On the whole, I think teachers and senior leaders are doing all we can to provide an excellent level of support and care to the students in our classes. But at the same time, I know many of you, like me, feel that despite these monumental efforts, we are not able to meet the needs of all pupils effectively.

We don't have the time, the energy or the resources to always perform at our best. Class sizes continue to rise. Post COVID, we've got an even greater number of complex needs amongst the students in our classes. And at the same time, classroom support is being cut due to budget constraints.

Consequently SEND and EAL provision is compromised and resources are often supplemented from teachers own salaries. It's not a pretty picture. Now in the spirit of full transparency, I'm very lucky that my school doesn't currently limit my use of the photocopier or demand that I achieve beyond expected progress with pupils who have such extreme needs that they can't access the curriculum.

But this podcast isn't just about my experience, and so I acknowledge that I'm one of the fortunate few, and I'm here trying to address the state of education as a whole. But even if you like me teach in a school or a college where you don't have to provide your own glue sticks and pencils, we're all human beings, which means that there will be days when we are on point and teaching excellent lessons.

And there'll be others where we are not able to deliver our real potential because we're having an off day. We fall ill, we can be swamped by our workload, or we're just not firing on all cylinders. And just as in every profession, there are going to be a limited number of outstanding teachers in any given school.

Many of those appointed will be hard working, solid practitioners. They're competent, but they may not be remembered by their colleagues and pupils as scarecrows. They're not outstanding in their field. I know, I know. I apologize, but we've just had Christmas and the cracker jokes are fresh on my mind.

We're flawed and we embody all the messiness that comes with being humans. And that is 100 percent fine. No one can ask us to give more than our best effort, but consider this scenario. Your son, your daughter, or someone that you cherish has to undergo a life saving operation.

It's a particularly tricky procedure, and you have a choice. You can roll the dice and be randomly assigned a human surgeon, or you can choose an AI robotic system with a proven track record, which has technical knowledge and dexterity that exceeds all but the best medical practitioners in the world. Is there really a decision to be made here?

But what if the rise of surgical robots meant that some human medics were let go? Should we have our loved ones face a poorer prognosis in order to protect their employment status? It's a question of utility and efficiency. We will always gravitate to a system or an approach that maximizes the outcomes we value.

Now I know my analogy isn't perfect and I appreciate that surgeons and teachers fulfill very different functions, but I'm raising the comparison precisely because our response to emerging technologies isn't always pragmatic or logical, but can be emotionally driven. If we look dispassionately at the use of AI in education, we should surely look for an approach that best serves the students in our care, even ahead of our own job security.

So let's lean into this thought experiment. Let's put the shoulds and the should nots to one side just for a moment and ask instead, Could artificial intelligence deliver a program of learning that is equal to or better than that provided by a teacher like me with all my foibles, inconsistencies and fluctuating quality of delivery, as well as my more inspired moments in the classroom?

Well, if we look at the current state of AI tools and the direction in which artificial intelligence is heading, I feel that the answer is almost certainly yes. There are already a number of AI infused platforms that are affordable and scalable. They're adaptive, they're highly personalized and they are calibrated for an optimized one to one learning experience.

A pupil logs in, And their previous interactions and metrics are called up and used to build a responsive coaching session paced to that pupil specific needs a combination of multimedia resources are used, including perhaps virtual or augmented reality simulations. To provide an immersive and stimulating lesson with immediate feedback and data tracking.

Unlike in a traditional classroom, the pupil is receiving ongoing one to one attention at all times. So their progress is likely to be accelerated in line with Bloom's two sigma problem. These learning models don't have sick days. They consistently deliver excellent lessons and have a level of subject expertise that is unparalleled.

And this is the worst they will ever be. Give it another decade and pupils probably won't even need to be staring at a screen. They'll be able to interact with a holographic avatar that's indistinguishable from a human being. In fact, they'll be able to probably choose their teacher's gender, age, accent, personality, and sense of humor.

So I'd probably be out for a start. With all this in mind. Should we expect to find ourselves in the unemployment line alongside all those surgeons? If we did shift to a teacherless model of education, what would be gained? And what would be lost? What's the cost benefit analysis here? The best thing we can do is look to real world case studies where this type of model has already been implemented.

And I'm going to focus on two today, Khan Academy, which is one of the most successful digital learning platforms in the world right now. And David Game College, which is a London based school. That's just started a pilot, a teacherless program for some of its year 10 pupils. I'm sure you've come across Khan Academy before.

It was launched in 2008, four years after the founder, Sal Khan, had been asked to tutor his 12 year old niece. And in doing so, he came to coach other friends and family members and noticed that there were certain patterns of limitation in the provision they had in their schools. So he devised some software to try and fill that gap.

It took off and it proved to be so successful that he quit his own job in order to give more in time and attention to the project. The mission statement for Khan Academy is to provide a free world class education, which is available to everyone, everywhere. And he had some very powerful sponsors and support from Google and the Gates Foundation who bankrolled his early efforts.

In the summer of 2022, Sam Altman from Open AI reached out and his team and the developers at Khan Academy explored how they might be able to use some of Open AI's artificial intelligence across the existing platform.

And the result was Khan Amigo. Their AI teaching assistant. Khan Academy currently enjoys 160 million users in 190 countries. So I think it's fair to say that it meets the conditions for proof of concept. In an interview with Ed Mylett, and again, I'll put a link in the show notes. Sal Khan is asked whether there will still be a need for bricks and mortar learning spaces where people come together to learn or does he think that the traditional model of education is going to disappear In the same way that blockbuster video evaporated in the face of streaming services like Netflix. 

I think you'll find Sal's response encouraging. He says that there will always be a place for good teachers and for physical centres of learning, but he imagines that the evolution of a kind of hybrid educational model is soon to come, where Students have access to personalised learning platforms like the ones I've just described, but then these sessions are augmented by human teachers, teachers who will then have the time and the space to be facilitators and mentors rather than just anxious grade chasers.

Arguably, this would offer the chance to return to a purer form of pedagogy, the kind of teaching that happened before we invested in the production line model of education, it could lend itself to an authentic Socratic series of interchanges. We've all heard the adage be a guide on the side rather than a sage on the stage.

But it's hard to embody that role sometimes when you're racing to get through the curriculum and teaching to the exam. Rather than sounding the death knell for teaching, this new vision could usher in a renaissance of sorts. By the way, if you want to learn more about Khan Academy and Sal's philosophy, I'd encourage you to pick up a copy of his book, 'Brave New Words, how AI will revolutionize education and why that's a good thing'. It's an excellent read. 

The second case study I'd like to explore today is an initiative launched at David Game College in London. In September last year, that's September 2024, they launched a year long trial for the first cohort of 20 students in their Sabre Wing program. It offers a  teacherless classroom experience on campus, where students engage with personalized learning pathways supported by learning coaches, but not subject specialists. And then these online sessions, which take place in the morning, are complemented in the afternoon with a curriculum of citizenship, critical thinking, digital literacy, public speaking, entrepreneurship, self awareness and artistic expression classes. The program's targeted for 15 to 17 year olds. And at the end of that 12 month period, students will come away with a minimum of five GCSEs. So they're currently offering maths, English, biology, chemistry, physics, and then a choice of either computer science, geography, or history.

The program's named after the Sabrewing Hummingbird, which is known for its resilience, its adaptivity, and its ability to fly independently over great distances. And I've reached out to the college in the hope that I might be able to interview some of their senior leaders about the initiative, and I'll let you know if this bears fruit.

My first impressions are that this may well be a model of how schools and colleges will have to adapt in the near future. The program would seem to me to offer the best of both worlds. There's independent study using cutting edge technology. But there's also professional guidance and support. There's an elective curriculum with some degree of freedom, and then there's a complementary co curricular program to ensure that pupils receive an holistic education.

There's accountability and structure. Provided to the students by being on campus and they engage in an active community with access to all of the resources that are available on a well provisioned campus such as a sports program, practical science experiments and art design facilities. But of course, at the moment, such a learning experience comes at a premium cost, and it's provided by an independent boarding school that can afford to test the waters as an early adopter.

I think it would be no mean feat for this to be replicated across the country in the maintained sector, as well as those independent schools who are now facing their own financial challenges with the loss of VAT exemption, which comes into effect this month. And this really brings me to my last point.

Even if we find that artificial intelligence is capable of augmenting the work that we do as teachers and that a hybrid model like this leads to the most efficient kind of educational delivery. It doesn't mean that we could implement it anytime soon because of the complexities of the schooling system that's currently in place, it would require a significant initial investment to upscale it so that it was fit for purpose.

In the provision of devices to pupils, and then there'd be the cost of the investment in the various digital learning platforms. Some of these costs could well be found by reducing the number of teachers on site. Our salaries are by necessity one of the biggest annual investments made by schools. But given the lead in required to make such a radical shift in provision, I think this is likely to take place gradually. We'll probably see more modest recruitment and a natural wastage through retirement as the new learning model is phased in over a number of years, rather than having to worry about a culling of teaching staff. I also think that such an approach is more likely to be embraced at secondary school level than it is with primary schools or in early learning environments where I think there's less scope for this kind of online self directed learning as the backbone of a curriculum. 

So in summary, I think we don't have much cause to be afraid that AI is coming for our jobs in the foreseeable future. I do think we face significant changes to the educational landscape and we will need to be adaptive and resilient in the light of these changes. We should bear in mind that artificial intelligence is already beginning to radically change the workplace and the skills and competencies sought by employers will shift to accommodate this.

The current standardized testing model of SATs, GCSEs, A levels and even diplomas and degrees may matter less to a prospective recruiter than a graduate's ability to show competence in using A. I. tools. To think critically and creatively. And this may well filter down to what we are asked to offer and deliver in our lessons.

The best way we can prepare in the light of all of this uncertainty is to become conversant with the AI tools that are available now. To stay abreast of the latest developments in AI, and then we'll be best positioned to not only future proof our own careers, but also help support our students to navigate whatever lies ahead.

Thank you for listening.  I hope you found some useful takeaways from this conversation. Please do spread the word if you think others would like the show and do check out the AI academia YouTube channel where you'll find practical tutorials that complement the topics covered on this podcast.

Have a great week. And I look forward to chatting again soon.