What this essay does is that it introduces a brand new education system. This education system is way more effective than the current education system we have in place. And unlike the current education system, which has been the same for a long time. This one is constantly getting better with time and data, thanks to AI.

The best thing about it is that it's totally free of cost. So anyone in the world with access to a basic smartphone+Internet can access Education 2.0. Meaning, now everyone has the same educational opportunities as a Harvard/Stanford student.

You see, education is the most fundamental thing to human progress. And once you deploy something like Education 2.0 on a global scale, it will have huge positive implications on economic growth, life quality, and human progress. And to emphasise what I just said, the collective intelligence of humanity is the best predictor of future progress. Here is some empirical data to prove this statement. Nevertheless, you don't need empirical evidence; look around, education is what drives societal growth.

Looking at the current education system we have in place. It would be unfair to say that it hasn't done an excellent job in aiding human progress. Resolving the COVID-19 Pandemic, the Apollo Program, the iPhone, Tesla, SpaceX, the Internet, All of Science etc. are a prime example.

But it certainly hasn't been able to keep up with the growing human population, and the challenges raised by it. Looking at all the pressing issues, we face today like global poverty, climate change, and war. The incapability of civilisation to solve these problems stems from the fact that it's not educated enough as a whole. There is conclusive evidence to prove this — Education correlates with prosocial behaviour. Prosocial behaviour is behaviour, which affects society as a whole.

So the best way to ensure that we progress into a brighter future is by rewiring the current education system. So it can keep up with the ever-growing population and the challenges we face, which will then eventually increase the collective wisdom and prosocial behaviour of humans at a much faster rate.

There are 3 main issues with the current education system, which dramatically lowers its potential.

  1. Its inability to reach a large population of the world.
  2. The unequal distribution of resources.
  3. The low efficacy of the current learning environments (i.e. lectures/books).

And Education 2.0 fixes all of these issues.

No access to education

The biggest problem with the current education system is that it's unable to reach an unfairly enormous number of the global population. People say that education is a human right, but looking at the current situation — that doesn't seem so.

According to Humanium.org, 72 million children of primary education age are not in school, and 759 million adults are illiterate. And we haven't even considered the inconsistency in educational outcomes yet.

Now I would like to explain the enormous implications this has on the economy. But first, I would like to establish a conclusive link between Education and GDP per capita.

Now doing the math:

  • $18,381 is the global average GDP per capita (a measure of economic output per person), and we have 831 million people with no access to education at all.
  • 831 million people * $18,381 ~ $15 Trillion worth of economic output wasted. And it has a growth rate of ~ 1.8% per year (Taking into account the growth rates of global GDP & global population).
  • To put this number in perspective, this is more than the GDP of ChinaIndia, and the UK separately, and $5 Trillion less than the USA.

There is no doubt that the current model of the education system is responsible for this issue. An education system is basically a collection of what we call "learning environment.". And a learning environment is an environment where learning takes place (i.e. classroom).

You see, the major problem with the current system is the high relative cost of setting up that learning environment (i.e. a classroom with qualified teachers, furniture, and students, etc.). So when you scale the education system, the setup + maintenance costs increase with it. And some countries just don't have the resources to scale. This can be due to many factors like large population, not enough teachers, etc. All of these issues then boil down to one primary issue — lack of economic resources, aka no money.

Scaling requires huge costs, which then dramatically lowers the potential for impact. Due to which, a significant amount of people in the world don't have access to basic education.

To beat this issue, what we need to do is switch to a new model of the education system. One which can scale cheaply and effectively. The best way to do this is by moving to a more digitalised education system.

One way to go about doing this is by leveraging Software, AI, and low-cost smartphones. Software and AI can be leveraged to create digital learning environments, which are more effective than the current learning environments we have, and its efficacy improves with time & data. These AI-taught lessons will be way better than attending a Stanford/Harvard lecture.

Once this works, the next thing to do is to ensure it supports low-cost smartphones, which will then allow for distribution at an enormous scale, while keeping the costs minimal. This will go a really long way in closing the educational gap. The best thing is that now anyone in the world will be able to have the same educational opportunities as an Ivy-League student.

In the long-term, you will see substantial positive implications on collective prosocial behaviour, economy, and quality of life. And redundancy in the current education system.

Unequal distribution of educational resources

First, we looked at inequality in terms of access to education. Now, we're going to look at inequality inside the people with access to education (in quality of educational resources). Turns out that this has huge negative implications on the overall learning outcome of the education system.

It is implausible to sustain the quality of educational resources (e.g. teachers, books, etc.) as you scale (make it available for a larger population) the education system. Due to which, a major side-effect of scaling is that it induces a widely unequal distribution of educational resources, which vary by demographic. Which then leads to variation in educational outcomes.

This is why the number of high-school graduates who attend prestigious colleges tends to be higher in private high-schools, relative to a public high-schools. The quality & quantity of educational resources (quality of teachers, study material, support, etc.) available to students at the private school is more elevated. Which then leads to higher success.

Inequality in educational resources leads to a variation in educational outcomes, which then lowers the average output of the education system.

Now, coming around to solve this issue. We need to create a system where the quality of educational resources is the best on earth. And it's available to everyone, no matter their economic situation. And finally, as it scales, its overall effectiveness doesn't reduce.

Low efficacy of current learning environments

The final issue I would like to talk about is regarding the teaching methods used currently. Teaching methods are ways in which teachers teach the student.

The problem is that the current teaching methods, which are used around the world, are scientifically flawed. This then makes the process of learning inefficient and ineffective. And on a large scale, it reduces the learning output (the product of learning) of the education system.

You might think that sitting in a classroom, while a teacher speaks out everything, and you note it down is an excellent way to learn. But in fact, this method is flawed. The reason it's flawed because it bears no consideration whatsoever to how our brain operates during learning. Due to which, the conversion rate from auditory + visual perception to long-term memory is low. This dramatically lowers the learning potential of the student.

The modern K-12 education system was established back in the 19th century to prepare the population to participate in the Industrial Revolution. At the time, not a lot of thought was given to the science of how we learn; they just wanted a cheap way to deliver information to a large number of students. And surprisingly, this factory model of the education system hasn't changed since then.

The starting point for this digital learning environment should be something, which is way more effective than any learning environment present today. Deploying this on a global scale will yield dramatically higher overall learning output than today. Making the current education systems redundant.

This would be the first step in accelerating the advent of Education 2.0. A world where high-quality education is free of cost to anyone and the quality of education is increasing over time. In the long term, this will mitigate all forms of inequality, and help humanity get onto a path of something bigger and brighter, thanks to its more immense collective wisdom.

Education 2.0 

We have now talked about the major issues our education system faces and potential solutions to them. I would now like to introduce a brand new education system — Education 2.0.

Introducing Education 2.0! It provides high-quality education, which is free of cost, and it is accessible to everyone in the world with a basic smartphone & internet connection. It ensures that everyone in the world has access to high-quality educational resources, not just a small sample from the vast population we have.

Making education free of cost for everyone is one of the core principles of Education 2.0. To ensure it's sustainable, it would have an advertising-based business model. But we would ensure that the advertisements don't affect the learning outcome of the system.

Once something like this is available — the next goal is to make it the single education system on the planet. Achieving this goal is incredibly important to mitigate educational inequality, which would then diminish all forms of inequality in the long-term.

The best way to go about achieving this goal is by convincing the entire world that Education 2.0 is better than the current education system they're enrolled in. And the best way to prove them is by deploying Education 2.0 to people who don't have access to any education system at all, as shown above — unfortunately, there are a large number of them.

We could then enrol them in our education system and run necessary scientific studies on learning outcomes. If Education 2.0 is built correctly — its learning outcomes should be significantly better than the current education system (at least 10X). This would go a long way in proving the effectiveness of Education 2.0. And getting others on board.

Thanks a lot to Charlie Liu for reviewing drafts of this essay.

New Comment
3 comments, sorted by Click to highlight new comments since:

I don't think this identifies, let alone addresses, any actual problems in "education". Quotes because it's such a big topic that I start out skeptical of anything that doesn't specify what elements it's addressing. You talk about K-12, and about Harvard/Stanford, without recognizing that there are at least dozens, likely thousands, of very distinct stakeholder and student-capability groups that require very different approaches.

My advice would be to decompose your recommendations. Find parts that address some specific need well, and that can be done in parallel with rather than in place of current systems.

The problem is that access to information/educational resources is almost never the problem. In poor countries, the best interventions to improve educational attainments are often health interventions. In rich countries, education isn't neglected, and it seems extremely hard to make real progress. Bill Gates recently spent almost a $1B to improve educational attainment in the US, with nothing to show for it.

Also, so far at least, ed-tech isn't looking too promising. Education can probably be improved with free online resources and maybe AI, but results have been disappointing thus far.

The larger issue is that instead of being a force for greater economic productivity and the creator of the public good of an educated citizenry, higher education in rich countries is largely a wasteful zero-sum competition for credentials. I recommend Bryan Caplan's book The Case Against Education on this topic (and Zvi's LW post on the book ). Improving education will therefore depend on changing credentialing.

For an analysis of the link between growth and education, I would recommend this post: Educational Romanticism & Economic Development .

I would also be very skeptical of claims that education causes prosocial behavior rather than just being correlated with it.

From an EA perspective, the current consensus is that education is not an effective cause area (unless "you are particularly well-suited to working in this field (e.g. you have lots of relevant knowledge or think you’d be a fantastic teacher), and you plan to do something innovative" ). See 5 reasons not to go into education and Schools Are Not the Key.

For such a long article about Education 2.0 it contains surprisingly little information about Education 2.0. All I found is that it's online, and free of cost (with advertising).

You know what else is already online and free of cost? Khan Academy, Wikipedia, and many educational YouTube channels. If "online" and "free of cost" were sufficient, the educational revolution would already be here (at least in English-speaking countries).

There are many online courses, many of them free, and yet despite the original enthusiasm, they didn't improve the situation much. Turns out, for most students, if there isn't someone checking their regular work, they simply won't work... they will procrastinate as long as possible, then try the first lesson, find out that it is difficult, and give up.

Ironically, a week ago we had here a great article about solving exactly this problem: DARPA Digital Tutor: Four Months to Total Technical Expertise?

So, what you propose is actually already here, and it didn't help. We can probably do it better, but you didn't address that part at all.