Homo Deus

by Yuval Noah Harari

  At a deeper level, as genetic technologies are integrated into daily life, and as people develop increasingly intimate relations with their DNA, the single self might blur even further, and the authentic inner voice might dissolve into a noisy crowd of genes. When I am faced by difficult dilemmas and decisions, I may stop searching for my inner voice, and instead consult my inner genetic parliament.

  On 14 May 2013 actress Angelina Jolie published an article in the New York Times about her decision to have a double mastectomy. Jolie lived for years under the shadow of breast cancer, as both her mother and grandmother died of it at a relatively early age. Jolie herself did a genetic test that proved she was carrying a dangerous mutation of the BRCA1 gene. According to recent statistical surveys, women carrying this mutation have an 87 per cent probability of developing breast cancer. Even though at the time Jolie did not have cancer, she decided to pre-empt the dreaded disease and have a double mastectomy. In the article Jolie explained that ‘I choose not to keep my story private because there are many women who do not know that they might be living under the shadow of cancer. It is my hope that they, too, will be able to get gene-tested, and that if they have a high risk they, too, will know that they have strong options.’28

  Deciding whether to undergo a mastectomy is a difficult and potentially fatal choice. Beyond the discomforts, dangers and financial costs of the operation and its follow-up treatments, the decision can have far-reaching effects on one’s health, body image, emotional well-being and relationships. Jolie’s choice, and the courage she showed in going public with it, caused a great stir and won her international acclaim and admiration. In particular, many hoped that the publicity would increase awareness of genetic medicine and its potential benefits.

  From a historical perspective, it is interesting to note the critical role algorithms played in this case. When Jolie had to take such an important decision about her life, she did not climb a mountaintop overlooking the ocean, watch the sun set into the waves and attempt to connect to her innermost feelings. Instead, she preferred to listen to her genes, whose voice manifested not in feelings but in numbers. Jolie felt no pain or discomfort whatsoever. Her feelings told her: ‘Relax, everything is perfectly fine.’ But the computer algorithms used by her doctors told a different story: ‘You don’t feel anything is wrong, but there is a time bomb ticking in your DNA. Do something about it – now!’

  Of course, Jolie’s emotions and unique personality played a key part too. If another woman with a different personality had discovered she was carrying the same genetic mutation, she might well have decided not to undergo a mastectomy. However – and here we enter the twilight zone – what if that other woman had discovered she carried not only the dangerous BRCA1 mutation, but another mutation in the (fictional) gene ABCD3, which impairs a brain area responsible for evaluating probabilities, thereby causing people to underestimate dangers? What if a statistician pointed out to this woman that her mother, grandmother and several other relatives all died young because they underestimated various health risks and failed to take precautionary measures?

  In all likelihood, you too will make important decisions about your health in the same way as Angelina Jolie. You will do a genetic test, a blood test or an fMRI; an algorithm will analyse your results on the basis of enormous statistical databases; and you will then accept the algorithm’s recommendation. This is not an apocalyptic scenario. The algorithms won’t revolt and enslave us. Rather, the algorithms will be so good in making decisions for us that it would be madness not to follow their advice.

  —

  Angelina Jolie’s first leading role was in the 1993 science-fiction action film Cyborg 2. She played Casella Reese, a cyborg developed in the year 2074 by Pinwheel Robotics for corporate espionage and assassination. Casella is programmed with human emotions, in order to blend better into human societies while pursuing her missions. When Casella discovers that Pinwheel Robotics not only controls her, but also intends to terminate her, she escapes and fights for her life and freedom. Cyborg 2 is a liberal fantasy about an individual fighting for liberty and privacy against global corporate octopuses.

  In her real life, Jolie preferred to sacrifice privacy and autonomy for health. A similar desire to improve human health may well cause most of us to willingly dismantle the barriers protecting our private spaces, and allow state bureaucracies and multinational corporations access to our innermost recesses. For instance, allowing Google to read our emails and follow our activities would make it possible for Google to alert us to brewing epidemics before they are noticed by traditional health services.

  How does the UK National Health Service know that a flu epidemic has erupted in London? By analysing the reports of thousands of doctors in hundreds of clinics. And how do all these doctors get the information? Well, when Mary wakes up one morning feeling a bit under the weather, she doesn’t run straight to her doctor. She waits a few hours, or even a day or two, hoping that a nice cup of tea with honey will do the trick. When things don’t improve, she makes an appointment with the doctor, goes to the clinic and describes the symptoms. The doctor types the data into the computer, and somebody up in NHS headquarters hopefully analyses this data together with reports streaming in from thousands of other doctors, concluding that flu is on the march. All this takes a lot of time.

  Google could do it in minutes. All it needs to do is monitor the words Londoners type in their emails and in Google’s search engine, and cross-reference them with a database of disease symptoms. Suppose on an average day the words ‘headache’, ‘fever’, ‘nausea’ and ‘sneezing’ appear 100,000 times in London emails and searches. If today the Google algorithm notices they appear 300,000 times, then bingo! We have a flu epidemic. There is no need to wait till Mary goes to her doctor. On the very first morning she woke up feeling a bit unwell, and before going to work she emailed a colleague, ‘I have a headache, but I’ll be there.’ That’s all Google needs.

  However, for Google to work its magic, Mary must allow Google not only to read her messages, but also to share the information with the health authorities. If Angelina Jolie was willing to sacrifice her privacy in order to raise awareness of breast cancer, why shouldn’t Mary make a similar sacrifice in order to fight epidemics?

  This isn’t a theoretical idea. In 2008 Google actually launched Google Flu Trends, that tracks flu outbreaks by monitoring Google searches. The service is still being developed, and due to privacy limitations it tracks only search words and allegedly avoids reading private emails. But it is already capable of ringing the flu alarm bells ten days before traditional health services.29

  A more ambitious project is called the Google Baseline Study. Google intends to build a mammoth database on human health, establishing the ‘perfect health’ profile. This will hopefully make it possible to identify even the smallest deviations from the baseline, thereby alerting people to burgeoning health problems such as cancer when they can be nipped in the bud. The Baseline Study dovetails with an entire line of products called Google Fit. These products will be incorporated into wearables such as clothes, bracelets, shoes and glasses, and will collect a never-ending stream of biometrical data. The idea is for Google Fit to feed the Baseline Study with the data it needs.30

  Yet companies such as Google want to go much deeper than wearables. The market for DNA testing is currently growing in leaps and bounds. One of its leaders is 23andMe, a private company founded by Anne Wojcicki, former wife of Google co-founder Sergey Brin. The name ‘23andMe’ refers to the twenty-three pairs of chromosomes that contain our genome, the message being that my chromosomes have a very special relationship with me. Anyone who can understand what the chromosomes are saying can tell you things about yourself that you never even suspected.

  If you want to know what, pay 23andMe a mere $99, and they will send you a small package with a tube. You spit into the tube, seal it and mail it to Mountain View, California. There the DNA in your saliva is read, and you receive the re
sults online. You get a list of the potential health hazards you face, and your genetic predisposition for more than ninety traits and conditions ranging from baldness to blindness. ‘Know thyself’ was never easier or cheaper. Since it is all based on statistics, the size of the company’s database is the key to making accurate predictions. Hence the first company to build a giant genetic database will provide customers with the best predictions, and will potentially corner the market. US biotech companies are increasingly worried that strict privacy laws in the USA combined with Chinese disregard for individual privacy may hand China the genetic market on a plate.

  If we connect all the dots, and if we give Google and its competitors free access to our biometric devices, to our DNA scans and to our medical records, we will get an all-knowing medical health service, which will not only fight epidemics, but will also shield us from cancer, heart attacks and Alzheimer’s. Yet with such a database at its disposal, Google could do far more. Imagine a system that, in the words of the famous Police song, watches every breath you take, every move you make and every bond you break. A system that monitors your bank account and your heartbeat, your sugar levels and your sexual escapades. It will definitely know you much better than you know yourself. The self-deceptions and self-delusions that trap people in bad relationships, wrong careers and harmful habits will not fool Google. Unlike the narrating self that controls us today, Google will not make decisions on the basis of cooked-up stories, and will not be misled by cognitive short cuts and the peak-end rule. Google will actually remember every step we took and every hand we shook.

  Many people will be happy to transfer much of their decision-making processes into the hands of such a system, or at least consult with it whenever they face important choices. Google will advise us which movie to see, where to go on holiday, what to study in college, which job offer to accept, and even whom to date and marry. ‘Listen, Google,’ I will say, ‘both John and Paul are courting me. I like both of them, but in a different way, and it’s so hard to make up my mind. Given everything you know, what do you advise me to do?’

  And Google will answer: ‘Well, I know you from the day you were born. I have read all your emails, recorded all your phone calls, and know your favourite films, your DNA and the entire history of your heart. I have exact data about each date you went on, and if you want, I can show you second-by-second graphs of your heart rate, blood pressure and sugar levels whenever you went on a date with John or Paul. If necessary, I can even provide you with accurate mathematical ranking of every sexual encounter you had with either of them. And naturally enough, I know them as well as I know you. Based on all this information, on my superb algorithms, and on decades’ worth of statistics about millions of relationships – I advise you to go with John, with an 87 per cent probability of being more satisfied with him in the long run.

  ‘Indeed, I know you so well that I also know you don’t like this answer. Paul is much more handsome than John, and because you give external appearances too much weight, you secretly wanted me to say “Paul”. Looks matter, of course; but not as much as you think. Your biochemical algorithms – which evolved tens of thousands of years ago in the African savannah – give looks a weight of 35 per cent in their overall rating of potential mates. My algorithms – which are based on the most up-to-date studies and statistics – say that looks have only a 14 per cent impact on the long-term success of romantic relationships. So, even though I took Paul’s looks into account, I still tell you that you would be better off with John.’31

  In exchange for such devoted counselling services, we will just have to give up the idea that humans are individuals, and that each human has a free will determining what’s good, what’s beautiful and what is the meaning of life. Humans will no longer be autonomous entities directed by the stories their narrating self invents. Instead, they will be integral parts of a huge global network.

  —

  Liberalism sanctifies the narrating self, and allows it to vote in the polling stations, in the supermarket and in the marriage market. For centuries this made good sense, because though the narrating self believed in all kinds of fictions and fantasies, no alternative system knew me better. Yet once we have a system that really does know me better, it will be foolhardy to leave authority in the hands of the narrating self.

  Liberal habits such as democratic elections will become obsolete, because Google will be able to represent even my own political opinions better than myself. When I stand behind the curtain in the polling booth, liberalism instructs me to consult my authentic self, and choose whichever party or candidate reflects my deepest desires. Yet the life sciences point out that when I stand there behind the curtain, I don’t really remember everything I felt and thought in the years since the last election. Moreover, I am bombarded by a barrage of propaganda, spin and random memories which might well distort my choices. Just as in Kahneman’s cold-water experiment, in politics too the narrating self follows the peak-end rule. It forgets the vast majority of events, remembers only a few extreme incidents and gives a wholly disproportional weight to recent happenings.

  For four long years I may repeatedly complain about the PM’s policies, telling myself and anyone willing to listen that he will be ‘the ruin of us all’. However, in the months prior to the elections the government cuts taxes and spends money generously. The ruling party hires the best copywriters to lead a brilliant campaign, with a well-balanced mixture of threats and promises that speak right to the fear centre in my brain. On the morning of the elections I wake up with a cold, which impacts my mental processes, and causes me to prefer security and stability over all other considerations. And voila! I send the man who will be ‘the ruin of us all’ back into office for another four years.

  I could have saved myself from such a fate if I only authorised Google to vote for me. Google wasn’t born yesterday, you know. Though it doesn’t ignore the recent tax cuts and the election promises, it also remembers what happened throughout the previous four years. It knows what my blood pressure was every time I read the morning newspapers, and how my dopamine level plummeted while I watched the evening news. Google will know how to screen the spin-doctors’ empty slogans. Google will also know that illness makes voters lean a bit more to the right than usual, and will compensate for this. Google will therefore be able to vote not according to my momentary state of mind, and not according to the fantasies of the narrating self, but rather according to the real feelings and interests of the collection of biochemical algorithms known as ‘I’.

  Naturally, Google will not always get it right. After all, these are all just probabilities. But if Google makes enough good decisions, people will grant it increasing authority. As time goes by, the databases will grow, the statistics will become more accurate, the algorithms will improve and the decisions will be even better. The system will never know me perfectly, and will never be infallible. But there is no need for that. Liberalism will collapse on the day the system knows me better than I know myself. Which is less difficult than it may sound, given that most people don’t really know themselves well.

  A recent study commissioned by Google’s nemesis – Facebook – has indicated that already today the Facebook algorithm is a better judge of human personalities and dispositions even than people’s friends, parents and spouses. The study was conducted on 86,220 volunteers who have a Facebook account and who completed a hundred-item personality questionnaire. The Facebook algorithm predicted the volunteers’ answers based on monitoring their Facebook Likes – which webpages, images and clips they tagged with the Like button. The more Likes, the more accurate the predictions. The algorithm’s predictions were compared with those of work colleagues, friends, family members and spouses. Amazingly, the algorithm needed a set of only ten Likes in order to outperform the predictions of work colleagues. It needed seventy Likes to outperform friends, 150 Likes to outperform family members and 300 Likes to outperform spouses. In other words, if you happen to have clicked 300 Lik
es on your Facebook account, the Facebook algorithm can predict your opinions and desires better than your husband or wife!

  Indeed, in some fields the Facebook algorithm did better than the person themself. Participants were asked to evaluate things such as their level of substance use or the size of their social networks. Their judgements were less accurate than those of the algorithm. The research concludes with the following prediction (made by the human authors of the article, not by the Facebook algorithm): ‘People might abandon their own psychological judgements and rely on computers when making important life decisions, such as choosing activities, career paths, or even romantic partners. It is possible that such data-driven decisions will improve people’s lives.’32

  On a more sinister note, the same study implies that in the next US presidential elections, Facebook could know not only the political opinions of tens of millions of Americans, but also who among them are the critical swing votes, and how these votes might be swung. Facebook could tell you that in Oklahoma the race between Republicans and Democrats is particularly close, Facebook could identify the 32,417 voters who still haven’t made up their mind, and Facebook could determine what each candidate needs to say in order to tip the balance. How could Facebook obtain this priceless political data? We provide it for free.

  In the high days of European imperialism, conquistadors and merchants bought entire islands and countries in exchange for coloured beads. In the twenty-first century our personal data is probably the most valuable resource most humans still have to offer, and we are giving it to the tech giants in exchange for email services and funny cat videos.