What should we be teaching our kids given the rise in AI, automation and genomics?

When you work in technology the real sign of middle age is being repeatedly asked by friends and family what we should all be teaching our kids given the emergence of AI, automation and genomics.

I hope that sharing my own thoughts on this doesn’t load too much pressure on my children but there is a lot of evidence that making a public statement of intent increases the chance that we follow through on those commitments – I’m hoping the act of writing this article will make me a better parent.

I think we increasingly ask this question with a belief that the education system, and to some extent we as parents, are focused on many of the wrong things if we want our children to flourish in a world of artificial superintelligence.

My concerns are somewhat exacerbated by the fact that I have observed an inverse correlation between the seniority of technologists and how much digital time they allow their own children. This is not limited to concerns regarding our kids, recently the designer of the Facebook ‘like’ button revealed that he has his assistant install parental control to stop him downloading apps on his iPhone.

However, I am not sure complete abolition is the answer – not least because we will all need to work alongside machines in some capacity to stay competitive, they are already virtual appendages, perhaps becoming physical ones in the future.

I believe the answer lies in helping our kids work with digital tools to hone specific skills so that they might leverage technology, including critical thinking, creativity and compassion.


Critical thinking (rather than retention of information)

Many of us are fixated on when the singularity (the point that machines become smarter than us humans on all dimensions of intelligence) will come, Googler Ray Kurzweil estimates that it will come in the year 2045.

That will certainly be an inflection point in human history that we should all think through but in the interim machines are already superintelligent on many axes of intelligence – including memory and processing speed.

But the education system rewards progress on those exact axes, for example the emphasis on learning by rote – memorising. It’s ironic that it’s often harder for computers to perfect the things we learn as toddlers, such as dexterity and walking, rather than those we focus on at school.

We need our kids to develop skills that help them leverage machine’s superhuman strengths. This is true of many professions too – many of us are shocked when doctors look up information online in a consultation. Do we really expect our doctors to recall everything they had learnt at medical school? Do we not want them to consider subsequent developments?

Instead of glamorising memory we need to place a greater emphasis on speed of learning, understanding context, adaptation and on how to frame the right question, either as a search query or indeed as a more complex algorithm. Machines (at least until the singularity) only ever act on human instructions, framing our questions is therefore our first opportunity to succeed or fail – ask a bad question and you will get a bad answer.

Similarly, we all need to remember that if you ask a biased question you will get a biased answer. It will be increasingly important for all of us to become adept at checking our sources of data, I hope that this will be a critical part of the curriculum.

I therefore try to congratulate great questioning from my kids, more so than when they find answers. I also try to flag sources of information, simplify context and point out the times that I believe services are trying to manipulate them – for example advertising.


Creativity (rather than efficiency)

Algorithms are occasionally generative but most often they are trained to complete a specific task, usually in a more efficient manner than a human operator. Those human roles that were most repetitive have been the fastest to be disrupted – weavers, book keepers, telegraph operators.

This rule of thumb applies looking forward too – instead of becoming book keepers or drivers (high repetition and therefore risk of disruption) I hope my kids find roles that are tougher to automate – perhaps as carers or entrepreneurs. And those that truly excel will harness machines to do repetitive tasks and save their efforts for more creative actions.

Creativity is the use of imagination or original ideas to create something – this will increasingly be the point of human value creation in any domain and the most differentiating skill.

One trap we fall into is to think of certain subjects as creative – for example, we hold up some subjects such as art and design as creative but many students only become proficient at copying those that they have observed (a task that machines will always win at).

In contrast, science and maths is often perceived to be uncreative but the greatest scientists shaped human history with huge creative leaps.  Newton discovered gravity when he saw a falling apple while thinking about the forces of nature – one of the greatest acts of human creativity.

Often those valuable leaps of creativity require base knowledge, Newton would certainly not have been equipped to ‘discover’ gravity without a phenomenal foundation in physics and maths. And as technology accelerates we will need to learn continuously in order to keep this foundation.

We can therefore no longer assume an early period of learning (for example at university) can then be capitalised upon by our kids indefinitely, instead we will need to learn continuously throughout our working lives. We need our kids to learn to love learning itself. We need to celebrate the learning journey rather than the destination.

I suspect our kids will need to continuously learn a foundation of knowledge and then learn how to be creative upon it. Somewhat controversially, I therefore often wonder how useful it is for our kids to learn languages, including computer code.

After all, Google has just announced headphones that will live translate speech into 40 languages, as neural nets improve I would expect translation to be close to perfect within a decade or so. Similarly, deep learning will likely automate the writing of code relatively quickly – while it will be useful to know how algorithms work I suspect most will be written by machine against a specific human (or eventually machine) query.

So unless our kids show incredible aptitude and are therefore capable of learning faster than their peers I suspect the value in learning languages and code will be more about understanding how the world works, which as the Newton example teaches us is key.

We should value learning the basics to this end but we should place a greater emphasis on helping our kids learn how to be creative.

The good news is digital tools can help – I often watch my son playing Minecraft. There’s a lot to like in his play – he creates his own worlds, researches other players’ approaches and iterates them and invents new tools. These skills may translate well into 3D modeling and design in the future.

However, I also observe him playing small games within Minecraft which show no obvious benefits to me – I’m therefore struggle to say if Minecraft is a good or bad thing for his development, it is likely both.

In truth I think most online services and games are both positive and negative in this way – the challenge is the useful learning and creating is often less fun than leaning back and mindlessly consuming content. I like that I create photos in Instagram but hate that I often check how many like each receives.

Perhaps there is a billion dollar opportunity for someone to develop a service that takes a subjective view on products and services and blocks the more mindless elements and therefore distractions – enabling us to focus on the positive.

Unfortunately that service is likely a long way off, so perhaps we have to help our kids understand the elements of services that might impart new skills and to help them get comfortable with being bored instead of becoming dependent on filling every snippet of time with a mindless distraction. After all, there is significant evidence that continuous partial attention lowers IQ.

It’s sobering that the author of Hooked: How to Build Habit-Forming Products, Nir Eyal, recently shared at a conference the lengths he goes to protect his own family. He has installed in his house an outlet timer connected to a router that cuts off access to the internet at a set time every day. “The idea is to remember that we are not powerless,” he said. “We are in control.”

I hope that I can help my kids love learning which in turn will give them the foundation to be creative.


Perhaps the last axis that the machines will supercede us will be ‘humanity’ itself but much of technology seems to be compromising our humanity.

While machines give us greater and greater leverage for impact we are in danger of being further and further removed from society. We have seen the effect of filter bubbles in recent elections. We are also socialising less –  12th-graders in 2015 were going out less often than eighth-graders did as recently as 2009, instead they are choosing to socialise on Snapchat and Instagram. When we do socialise it tends to be more often with people just like us, perhaps this is creating an empathy gap of sorts.

I wonder if genomics will widen that gap further – in a world where a subset of society is able to genetically engineer their own offspring will they lose empathy for society writ large?

The last and perhaps most important skill I hope we might instill in our kids is empathy and humanity. This may also be the toughest to execute given that confirmation bias pulls us into homogeneous bubbles but perhaps we can harness digital technologies to weigh up the consequences of our actions and to empathise with others different from ourselves.

There is surely an opportunity for someone to design a digital product that exposes us to viewpoints different to our own, our challenge will be to fight the urge to discount those different opinions.

One experience that gives me hope is that my kids have stood in the shoes of a Syrian refugee through an incredible New York Times Google Cardboard virtual reality product, they felt they had been transported to the camp. In the future could they attend school for a day in the developing world?

Carl Rogers said there are as many different perceptions as there are people in the world, perhaps new tools will help us truly perceive the world as others do rather than just stand in their shoes.

Until then ensuring our kids spend time with people as different to them as possible will help foster the empathy they will need.

Our kids will stay in front of the machines and leverage them for good if we double down on our own humanity and continuously learn to create…