The Daily Telegraph - Saturday - Review
10 GIANT LEAPS: A TECH TIMELINE
1957
Sony TR-63
The trend for personal tech began with Sony’s pocket-sized transistor radio, which replaced bulky cathode sets.
1994
IBM Simon Personal Communicator
The grandfather of smartphones: calls, emails and faxes all on one touchscreen handset. 1998
Larry Page and Sergey Brin’s search engine has made them (respectively) the 6th and 8th richest people in the world.
1971
Intel’s first microchip Federico Faggin’s revolutionary design squeezed 2,250 processors into just 12 sq mm of silicon.
1996
Nokia 8110
This banana-shaped phone became a symbol of cutting-edge cool after a version appeared in the 1999 film The Matrix.
with your existing systems – open Amazon Storytime, play music from your Sonos library, reorder your cat food, turn on the Nest thermostat controls, and search the web multiple times faster than you can. An AI PA will be a mini-me – a neural network (a series of algorithms trained to recognise patterns) that learns my wants and needs, my favourite foods, my travel preferences, restaurants, calls, bills I forget to pay, birthdays I forget to remember, and all my digital photographs, my texts, my emails. My self, wherever I have hidden it.
And what about my politics? My dirty secrets? Even my thoughts? Here’s Larry Page on how Google will go in the near future:
Our ultimate ambition is to transform the overall Google experience, making it beautifully simple. Almost automagical, because we know what you want and can deliver it instantly.
And while what you want is being known and delivered, all of that will be tracked. Cookies, remember, are bits of tracking code inserted into your computer. When you, personally, are no longer doing the searching, because your personal PA is doing it for you, there will be no effective privacy preference.
But in fact there really isn’t one now. Even innocent-seeming apps, such as those for the weather or ride-sharing, are infested with tracking code. A minime PA will be a seductive choice. Why wouldn’t we want an able, considerate, smart helper who is always available, and mostly free? That used to be called a wife. But then feminism spoiled the party.
She or he could, of course, in time, be a double agent. In a Blade Runner world, I could be turned in to the authorities by my own virtual mini-me. And she’ll know where the money is. Where the bodies are. Who my friends are, and how to find them.
Can I run away? In a cashless world I will be using my phone to pay for everything at first – and then iris recognition, or fingerprints, or chip implant will do away with the need for external devices. I will be my own device. No need for a wallet, or a phone, or a set of keys, or an office-swipe. I will be free. And followed everywhere. Or, rather, there will be no need to follow me because my location will be obvious.
Sci-fi. Usually a dystopia for the hero and friends, and a utopia to those who benefit from or accept the situation. I suspect that the future will not be so binary. What’s for sure is that “privacy is an anachronism” (thank you, Mark Zuckerberg). Like every other system in the coming world, I too will always be on, always be known, always be available, even while I sleep, dream, or think about things. Soon the interface between me and mini-me – my Google self and myself – will be redundant. We will merge.
Think about a world where there are no private thoughts, no private actions. The internet of things will allow any object to act as a computer. Your fridge will tally the food you buy and eat. If you sign up to a dieting app, the fridge will “help” you by directly ordering the food you should be eating. The fridge will also self-lock if you break the rules. Desperate folks trying to hack their own fridge is a new torture coming your way soon. Smart beds will be able to monitor – and assist – a good night’s sleep, warming or cooling the bed, managing light flow, and reporting your state to your automated doctor; you might need medication. You might not be fit to drive today. Did you have sex? No? Is your relationship healthy? Perhaps you need a counsellor, or Viagra. In the smart kitchen, the toaster will remind you that today is a no-carbs day. The smart toilet will assess the contents of your initial evacuation. (I am not making this up.)
Leaving aside the monetisation of every breath you take, there will be advantages to being seamlessly connected. Chore-work and bore-work can be taken care of – who wants to go to a supermarket, report the faulty boiler, wait in for the plumber, or manage their health, when check-ups with the GP can be made by your monitoring implants, and smart homes will run themselves, including taking receipt of goods, and letting in the plumber, whose movements will be visible on your phone and whose access will be strictly timed? When you are gaining so much, does it matter if there are no secrets, and perhaps no self, anymore?
The new reality will not be sold to us as surveillance, with its totalitarian overtones. The future will be sold to us as empowerment. Elon Musk’s Neuralink company is working on brain-computer interfaces – threads that will allow someone to control a computer via their thoughts. Human trials started in 2020. The current aim of the tech is to help people with paralysis, a laudable aim. Musk’s eventual aim, though, seems to be symbiosis with AI, so that humans don’t get left behind in the intelligence game.
Modern medicine has already reset human biology. We live twice as long as our ancestors at the start of the Industrial Revolution. The rich, who have access to the best of everything, are doing very well. Naturally, they want to do better, which is why Silicon Valley is investing in research that will stall, or reverse, physical and cognitive decline.
Cognitive decline for humans is as real as muscle loss and organ failure. In the end, our biology beats us. AI systems, embodied or not, suffer no such losses and no such decline. AI systems can augment, version-up, get smarter. If humans are, say, HomoSapiens3, the post-Industrial Revolution version, we shall have to get to HS4 pretty soon if we want to stay in the game. Merging with the AI we are developing is a logical outcome. Outcomes, though, often resist prediction.
What can be predicted is personalisation. As we saw, personalisation began with that first transistor radio back in the 1960s: at last, a small portable device just for you – no need to sit round the family radio. Go your own way.
Personalisation was enthusiastically adopted by the laptop and smartphone industry. Simultaneous with our fully public-data-harvested-known selves is the personalisation of that self. It will be “your” smart implant. “Your” smart car/house/lifestyle/ insurance/portfolio/ personal shopper/fitness guru/ therapist/PA. Tailored to you, your tracker-helpers will change and develop seamlessly (frictionlessly) as you do.
That’s a clever marketing move. Personalisation isn’t just about the product anymore. It’s the concept. Personalisation is being offered in place of the oldfashioned, outmoded idea of privacy.
But why is privacy problematic in our internet-ofeverything future? Privacy is friction. In economicsspeak, friction is the opposite of flow. Friction is whatever impedes the data flowing from you and all that you do, to the interested parties who want to make money out of you, and/or control/nudge your behaviour. It is as simple as that.
Or am I just an analogue human who likes the idea of being off-grid sometimes? At present that is still possible if you leave your phone at home, walk wherever you want to go, pay cash where there is no CCTV (both increasingly difficult moves, I admit), don’t browse the internet for a few days. Soon, though, as smart devices and smart implants become normal, you won’t be logging in. You’re in. You’re on. For life. From sci-fi to Wi-Fi to my-wi.
Former CEO of Google Eric Schmidt, sitting next to Facebook COO Sheryl Sandberg at Davos in 2015, put it like this:
The internet will disappear. There will be so many IP addresses, there will be so many devices, sensors, things you are wearing, things you are interacting with, that you won’t even sense it. It will be part of your presence all the time.
For young people – digital natives who have grown up with a phone and Facebook, whose every move is on Insta, and who want to be influencers themselves – questions about protection, privacy, contentscreening and platform responsibility are being asked by an NGO called 5Rights. 5Rights was founded in the UK in 2015 by filmmaker Beeban Kidron. When I asked her about 5Rights, she pointed out that over a billion underage young people are online every day, for hours every day, and treated by the platforms they use as if they are adults.
Content is easy to access, hard to monitor. Online grooming is a particular threat. For example, during the 2020 Covid lockdown, in the UK (just the UK here, folks) around 9 million attempts to view childabuse images were blocked in one month alone.
Kids are tech savvy but tech vulnerable. 5Rights wants to see kids protected in the digital world just as they are in the physical world. In the physical world we do make a distinction between children and adults – a distinction hard-won during the Industrial Revolution. We don’t want our kids working in sweatshops, but we seem unconcerned about exploitation via their phones. That includes addictive gaming and porn habits, as well as the suicidal misery of “likes”.
Data collection that starts early in life amounts to a conquest of that life. And, as we have seen in the stand-off between China and Hong Kong, forced data removal from popular sharing sites like TikTok can be used to persecute or prosecute young people, to monitor their behaviour, and no doubt to influence their political “choices” later. In China’s case the datasnatch is clearly political. That’s not the point though. China is doing in an obvious way what is being done quietly and covertly in the “free” Western world every day. Our data is not anonymous. Who has a “right” to this data? To sell it? To snatch it? To package it?
What happens, though, when there is less, or even no, division between the online worlds and the physical world? When Eric Schmidt’s prediction of the end of the internet and the start of the internet of everything happens? How do we protect anyone when the internet is always on and we are always on it? I loved that great story of the online-addict teenager whose mother confiscated all her devices and turned off the Wi-Fi. The kid realised she could send tweets from the home’s smart fridge. The whole thing may well have been a hoax, but Reddit offers full instructions on how to tweet from a Samsung fridge, if you have one. The point of this story is that the goal of our digital masters in Silicon Valley is that there will be no offline. No need to hack your fridge. And yet…
It may be that all of this – privacy, data usage – is a temporary problem. At present we imagine human interests and human actors as dominant in all our scenarios. If, though, AI does become superintelligent – a player and not just a tool – then the future for humans may be irrelevant. I mean, how much data will AI need on a species being consigned to the Museum of History?
When I talk to people about the future, many believe that the world’s head-in-sand attitude to climate catastrophe will make that other kind of sand – silicon – irrelevant, in or out of a valley. We’ll be fighting for food, not tweeting from our fridges. Others believe that developing super-intelligent AI, as swiftly as we can, is our best chance of survival. Until 2020, none of us was thinking about viruses as the wipe-out call. Now we are.
Ironically, although the world may become much poorer because of Covid-19, the virus is a chance for the tech giants to get much richer, and to get more control. And not just Amazon doing home delivery. Eric Schmidt has been talking enthusiastically about homeschooling for all (except the rich, you can be sure), using the platforms that have begun to replace contact during the pandemic. If we are at home, we will need to be connected in new ways. That’s an opportunity for the connectors. Virtual-reality avatar sessions are being trialled by Facebook. I am sure they will become as real for us as Zoom is now.
More worryingly, track and trace everywhere, including even a visit to the pub, is going to allow levels of surveillance that civil-liberties groups would have taken years over, arguing about privacy and usage. That’s all gone now. Being watched equals being safe.
But what about energy constraints? AI is an energy guzzler, and even if we extracted all the fossil fuel left on the planet it wouldn’t be enough for the kind of super-future envisioned by Ray Kurzweil or Elon Musk. That’s why the premise of The Matrix is that humans are just battery packs in an AI simulation.
Optimists say that the energy requirements of an AI future will force the world towards low-carbon solutions. The market will drive the change because it must.
But there are other constraints on an AI future too. Intel co-founder Gordon Moore came up with his own law: Moore’s law observes that every two years the number of transistors that can fit onto a square inch of microchip will double. Fifty years after that first Intel chip, the computing power that once filled a whole building with hardware now fits into your handbag. And it uses much less power. That’s progress.
There is a limit to progress though – and unless we switch our systems we are pretty much at the limit. Put simply, there’s no more room, at the small scale of the laptop or phone, to keep doubling the number of transistors. No matter how tiny, they still take up (some) physical space.
The next jump to faster processing speeds and more commands will be quantum computing. There’s a rumour that China has made the breakthrough already. If they have, they aren’t telling anybody. Google and IBM are both claiming to be nano-close.
Transistors work on the familiar zero-and-one principle – whether it’s analogue or digital. “Bits” of information each hold a 1 or a 0. A quantum “bit”, or qubit, is different. Very different. By harnessing subatomic weirdness, a qubit can be a 0 and a 1 at the same time.
Numbers-wise, 8 bits make a byte. Your smartphone memory could have two gigabytes – that’s 2 x 8 billion bits – but a few dozen qubits is way, way beyond that. According to Dario Gil, director of IBM’s research unit in Yorktown Heights, NY:
Imagine you had 100 perfect qubits. You would need to devote every atom of Planet Earth to store bits to describe the state of that quantum computer. By the time you had 280 perfect qubits you would need every atom in the universe to store all the ones and zeros.
At present, IBM Q System One lives like a reclusive rock star inside a 9-foot cube of black glass only accessible through 700 lb doors a half-inch thick. Quantum computers must be absolutely remote from any entanglement with reality – entanglement affects the outcome, as anyone who has ever fallen in love knows. So, we are building a god so remote that it must live in an inaccessible temple visited only by high priests in special clothes. The high priests can ask the questions and interpret the answers. Quantum computing may be the future, but that story is like a Pharaoh-dream from the past.
Where is all this heading? What’s for sure is that fewer and fewer people will know how the systems that control us actually work. We’re not talking here about fixing the washing machine. While there is no consensus on the future, there is consensus that total connectivity will happen – to the internet, to our devices, to our machines, to each other. When you customise your connectivity it will feel like yours. Actually, it will feel like you. And you will feel as if you have chosen it. An avatar Sinatra singing “I did it my-wi”. And a whole new questioning of what is “you” comes bubbling up.
My-wi is religious in its own way. Mark Zuckerberg has talked about Facebook as a “global church”, connecting people to something bigger than themselves. It may be bigger – it may be smaller – but it will be connected. Here’s George Orwell in his novel Nineteen Eighty-Four:
You had to live – did live, from habit that became instinct – in the assumption that every sound you made was overheard, and, except in darkness, every movement scrutinised.
The success story of Homo sapiens has been one of infinite adaptability. Adapting ourselves to the Machine Age was an unprecedented rip with our evolutionary past. We mourn the price to Planet Earth, but few would wish to return to a pre-1800 world. We dislike the Intrusion of Everything, but who would want a world without smartphones and Google? Perhaps, though, we would prefer a world that may be less democratic, but that could also be less stressful for the next stage of our development.
My-wi might leave us as little children are: cared for, fed, safe, watched over, with plenty of fun stuff and free stuff, and with someone else deciding the big stuff. No reason to believe that who decides will always be in human form.
Computing power that filled a building with hardware now fits into your handbag