Are algorithms ruling your world?
By Ad astra
A year or two ago, how many would have known what the word algorithm meant? Now it is a word in common use. It crops up whenever automation or artificial intelligence is mentioned.
The term ‘automation’ once conjured up images of robots doing manual tasks; now it encompasses intellectual or cognitive tasks being undertaken automatically. We are told that already the majority of financial transactions are carried out not with pencil and paper and calculators, but via algorithms.
The images of robots scurrying round the factory floor building motor vehicles or fulfilling customer orders in a vast warehouse, as happens in the Amazon organization, are easy enough to envisage and understand, although the programming behind these activities would be a mystery to most of us.
How many understand how an algorithm works, or even what is it?
Although the concept of an algorithm dates back to the 9th Century, it has come into its own during this century as we seek to automate a multitude of tasks previously done manually.
A simple definition of an algorithm is a self-contained sequence of actions to be performed, beginning with inputs and finishing with outputs.
In computer parlance, an algorithm is a well-defined procedure, a sequence of unambiguous instructions that allows a computer to solve a problem. Algorithms can perform calculations, data processing and automated reasoning tasks.
The most familiar algorithm is a kitchen recipe. It comprises the ingredients (the inputs) and the directions, instructions about how to combine the ingredients to produce the dish (the output).
Wikipedia provides an example of a simple algorithm in mathematics – a set of instructions to find the largest number in a list of numbers arranged in random order. Finding the solution requires looking at every number in the list. This simple algorithm, stated in words, reads:
- If there are no numbers in the set then there is no highest number.
- Assume the first number in the set is the largest number in the set.
- For each remaining number in the set: if this number is larger than the current largest number, consider this number to be the largest number in the set.
- When there are no numbers left in the set to iterate over, consider the current largest number to be the largest number of the set.
For computer processing, those instructions are written in computer language, for example using ‘if – then – else’ propositions: IF ‘such and such is so’ THEN ‘do this’, ELSE ‘do this’.
Such straightforward mathematical algorithms seem harmless enough. An input is processed and the output is reliably produced.
Now though these mathematical calculations are used in commerce and finance, for example in stock market transactions where the computer programs of stock brokers compete with one another to accomplish the most advantageous transactions for their clients. There are stories of stock broking firms using faster and faster computers and building faster and faster transmission lines to the stock exchange to outdo their competitors. Even a few thousandths of a second faster transmission can make all the difference.
At times the speed and number of such competing automated instructions have brought the stock market to a halt – the ‘flash crash’.
We need though to get away from the notion that mathematical algorithms are pure and free from bias because they use the science of mathematics. Cathy O’Neil, a Harvard PhD and data scientist tells us why in her recently published book: Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy, published by Crown. Her experience working for a hedge fund at the time of the global financial crisis informs her views and her writings.
In an article in The Guardian last October by Mona Chalabi titled Weapons of Math Destruction: Cathy O’Neil adds up the damage of algorithms, Chalabi points out that algorithms that started as rule-based processes for solving mathematical problems are now being applied to more and more areas of our lives. She continues:
This idea is at the heart of O’Neil’s thinking on why algorithms can be so harmful. In theory, mathematics is neutral – two plus two equals four regardless of what anyone wishes the answer were. But in practice, mathematical algorithms can be formulated and tweaked based on powerful interests.
O’Neil saw those interests first hand when she was a quantitative analyst on Wall Street. Starting in 2007, she spent four years in finance, two of them working for a hedge fund. There she saw the use of weapons of math destruction, a term O’Neil uses to describe “algorithms that are important, secret and destructive”.
The algorithms that ultimately caused the financial crisis met all of those criteria – they affected large numbers of people, were entirely opaque and destroyed lives. O’Neil left the hedge fund: “I left disgusted by finance because I thought of it as a rigged system and it was rigged for the insiders; I was ashamed by that – as a mathematician I love math and I think math is a tool for good.”
According to O’Neil, algorithms can be used to reinforce discrimination and widen inequality, by ‘using people’s fear and trust of mathematics to prevent them from asking questions.’ This can occur when aspects of life other than objective mathematical propositions are the inputs to the algorithm.
Her book explains how algorithms can do this – such as the ones used to measure the likelihood a convicted person will relapse into criminal behaviour: ‘When someone is classed as “high risk”, they’re more likely to get a longer sentence and find it harder to find a job when they eventually do get out. That person is then more likely to commit another crime, and so the model looks like it got it right.’
O’Neil tells us that ’…contrary to popular opinion that algorithms are purely objective, “models are opinions embedded in mathematics”. Think Trump is hopeless? That will affect your calculations. Think black American men are all criminal thugs? That affects the models being used in the criminal justice system.’
But O’Neill tells us that sometimes it’s hard for non-statisticians to know which questions to ask. Her advice is to be persistent. ‘People should feel more entitled to push back and ask for evidence, but they seem to fold a little too quickly when they’re told that it’s complicated.’ She adds: ‘If someone feels that some formula has affected their lives, at the very least they should be asking, how do you know that this is legal? That it isn’t discriminatory?’
Algorithms have the capability to sort through vast amounts of data – so-called big data. But what data should algorithms be sorting?
Writing in The Guardian, in a article titled: How algorithms rule the world, Leo Hickman says: ‘From dating websites and City trading floors, through to online retailing and internet searches (Google’s search algorithm is now a more closely guarded commercial secret than the recipe for Coca-Cola), algorithms are increasingly determining our collective futures. Bank approvals, store cards, job matches and more all run on similar principles. The algorithm is the god from the machine powering them all, for good or ill.’
We are becoming aware that our Internet browsing history, our Google searches, our contributions to Facebook, Twitter and other social media are being monitored and fed back to us in the form of suggestions about what we might buy or eat or how we should vote.
The values and the objectives of those who design algorithms are reflected in the data collected and the algorithms used to process the data.
In The Guardian article: How algorithms rule the world Viktor Mayer-Schönberger, professor of internet governance and regulation at the Oxford Internet Institute, warns against humans seeing causation when an algorithm identifies a correlation in vast swaths of data.
He cautions us about: ‘… the possibility of using big-data predictions about people to judge and punish them even before they’ve acted. Doing this negates ideas of fairness, justice and free will. In addition to privacy and propensity, there is a third danger. We risk falling victim to a dictatorship of data, whereby we fetishise the information, the output of our analyses, and end up misusing it. Handled responsibly, big data is a useful tool of rational decision-making. Wielded unwisely, it can become an instrument of the powerful, who may turn it into a source of repression, either by simply frustrating customers and employees or, worse, by harming citizens.’
Mayer-Schönberger presents two very different real-life scenarios to illustrate how algorithms are being used. He explains how the analytics team working for US retailer Target can now calculate whether a woman is pregnant and, if so, when she is due to give birth: ‘They noticed that these women bought lots of unscented lotion at around the third month of pregnancy, and that a few weeks later they tended to purchase supplements such as magnesium, calcium and zinc. The team ultimately uncovered around two-dozen products that, used as proxies, enabled the company to calculate a “pregnancy prediction” score for every customer who paid with a credit card or used a loyalty card or mailed coupons. The correlations even let the retailer estimate the due date within a narrow range, so it could send relevant coupons for each stage of the pregnancy.’
‘Harmless targeting, some might argue. But what happens, as has already reportedly occurred, when a father is mistakenly sent nappy discount vouchers instead of his teenage daughter whom a retailer has identified is pregnant before her own father knows?’
Mayer-Schönberger’s second example of our reliance upon algorithms throws up even more potential dilemmas and pitfalls: ‘Parole boards in more than half of all US states use predictions founded on data analysis as a factor in deciding whether to release somebody from prison or to keep him incarcerated.’
Awareness of the useful potential of algorithms is valuable, but so is their propensity for doing harm in the wrong hands or for the wrong reasons. But how many of our citizens are aware?
Will we all awake one day and find that our lives are being controlled secretly by forces whose self interest, not ours, is being served? By forces that want us to buy in a certain way, transact our business in a certain way, view cultural and travel offerings in a certain way, vote in a certain way, behave in a certain way, and even think in a certain way? By forces that selectively benefit those at the top and penalize those at the bottom? By forces that increase the inequality that afflicts the world today?
Does that sound like George Orwell’s Nineteen-Eighty Four?
We ought to be afraid. We ought to resist strongly our lives being taken over and controlled by algorithms.
But who will listen? Are our politicians aware of the threat of algorithms? More significantly, are they capable of altering this surreptitious take over of our world?
Do you use Google, Bing or Yahoo as online search engines? Do you respond to emailed requests to take a quiz or respond to an online opinion poll? Are you attracted by offers of a prize if you respond to a survey? Do you enter contests that promise alluring rewards? Do you use YouTube or Netflix or Stan? Have you used iTunes or Google Play or Amazon online to order items?
Have you noticed that your online searches often mysteriously throw up the very things that interest you?
If so, chances are that you may already be in the thrall of the algorithm creators, already slaves to the algorithm.
Are algorithms ruling your world?
What is your opinion?
Do you feel you are being manipulated through your Internet searches?
Have you had any troublesome experiences using the Internet?
This article was originally published on The Political Sword
For Facebook users, The Political Sword has a Facebook page:
Putting politicians and commentators to the verbal sword
Like what we do at The AIMN?
You’ll like it even more knowing that your donation will help us to keep up the good fight.
Chuck in a few bucks and see just how far it goes!
16 comments
Login here Register here-
Miriam English -
diannaart -
glenn k -
astra5 -
Harquebus -
king1394 -
diannaart -
astra5 -
astra5 -
astra5 -
diannaart -
jimhaz -
Harquebus -
Hupple -
astra5 -
Miriam English
Return to home pageI’ve been a computer programmer since before personal computers, so I am one of those who has well understood the power and utility of algorithms, along with the dangers of badly designed ones.
Many of the dangers presented in the above article are real, but are being repaired. The accidental racism and classism in the paroling algorithms, for instance, is being removed. Google puts an awful lot of effort into trying to ensure its algorithms don’t mess with people, but instead deliver to them what they actually want — and in a way, that is the greatest danger. Algorithms are helping people enclose themselves inside filter bubbles, where they only see what they want to see. I use this to my advantage. If I want Google’s search engine to look for the kind of thing it knows I want then I use it, but if I want to show up unusual things that might run counter to my preconceptions then I search using DuckDuckGo instead, because it purposely does not gather any personal information on preferences.
At the moment algorithms have been becoming more and more important, but I seriously doubt that will last very long. We are moving into the age of artificial intelligence (AI). In a sense AI is yet another system of algorithms, but they are only so in the way that our own minds are algorithms. AIs are genuinely intelligent machines that will bring great benefits to us all. There will be some hiccups along the way, of course, as with all new technologies, but our opportunities to partner with these AIs and each become smarter and have more control over the world we live in should not be understated.
Do I sound like a Pollyanna in thrall to technology? Yeah, I know it sounds like it, but in truth I’m acutely aware of the dangers of all technology, and although there are some dangers that can result (in the short term) from AI, the advantages to us all are so truly world-changing that AI is something we should grasp with both hands. We have benefitted greatly from personal computers. They have opened enormous vistas of information to us, made it possible to communicate on a scale undreamed of just a single human lifetime ago, given us all mathematical, grammatic, artistic, and research skills that we could never have hoped for. But this is the barest inkling of the way personal AIs will empower us all.
Because Alzheimer’s runs in the women in my family I fully expect to begin to feel its effects soon. I’m hoping I can have a personal AI to help me even when I’m deep in the clutches of that horrible disease. But “normal” people will even more greatly. They will become, in effect, far smarter than the most brilliant geniuses of history due to their personal AIs. And as AIs continue to grow in intelligence they will take us to ever greater heights. I wish I could live for another century to see it. Very exciting times ahead.
Wonderful article – do not trust algorithms – they are only as useful as the programmer (in oldspeak) or Coder (in newspeak) made them.
Even ones which are supposed to self-learn – it all boils down to what data goes in.
Do I feel I am being manipulated through Internet searches?
No, just annoyed.
For example, am almost regretting buying a new vacuum cleaner, as I am now pursued by waving tentacled… – apologies that was my last night’s nightmare… my point is even though new data has occurred, “have bought new cleaner” this never gets added to the equation. I continue to be plagued by vacuum cleaners.
I believe, therefore, algorithms devised only by teenage boys… which is mostly happening now – another reason for gender/age/ethnic equality/diversity.
Have I had any troublesome experiences using the Internet?
Being pursued by waving hoses is troublesome.
confirmation the stock market is rigged from the inside. has an investment bank ever reported a quarterly loss?
Miriam English
I thank you for your very insightful comment. You paint an exciting future for AI. I hope that you, indeed all of us, can benefit from it.
diiannaart
Thank you for your generous compliment,
Your experience with your vacuum cleaner exemplifies one of the problems occasioned by algorithms.
glen k
You are right about the stock market. I’m currently reading ‘Game of Mates – How favours bleed the nation’ – very revealing!
Bots are now infiltrating fora and twitter etc. I sometimes wonder if some of the conversations here are bot vs bot.
Sumpin’ I jus’ red.
“research was first hinting, and then confirming, that most of the clicks the ads were getting — that indicated their success, and for which they were charged money — were coming not from people but from bots, little automated apps that were manipulating data and getting into mischief but have never, ever, bought a product. ”
A good part of my library is devoted to algorithms. I have been programming and problem solving using sequence, selection and iteration for over 30 years. Luv it.
If you use the net, you will get advertisements. I’d rather have ads that reflect my interests than ads that stereotype me. A constant barrage of miracle cures for wrinkles pursues me, presumably because the system knows I’m an older female. I quite enjoy seeing the response to a single enquiry – recently I looked for a piece of sheet music and now I have lots of ads for ‘printable music’. Today I checked up on a medical illness being suffered by a friend, so I could have better understanding of what she is going through. I’m sure there will be a plethora of ads on health now.
Does it matter? I don’t think so. No one is making you look at the ads or buy the products – and as soon as I pursue some other subject, it will change the ads I receive
Since the great vacuum cleaner stalking, I now use Duck-duck-go – have fewer problems with pesky adverts.
Now, having just read through the comments, am considering the implications of an IT coder, trapped in a narrow range of thinking, writing algorithms, subconsciously self selecting for his/her bias.
Harquebus
What an interesting article you have linked us to. The comments that followed the article too were informative. Thank you.
From your experience over a lifetime you know that there is nothing inherently wrong with algorithms. They are invaluable tools for accomplishing a vast array of tasks. And will continue to be.
Problems that emanate from the use of algorithms are the result of their misuse, abuse, and exploitation by those who seek to gain a surreptitious advantage over those who do not know they are being manipulated. After reading your linked piece: Digital Advertising: The Rise and Fall of Crappy Crap, it seems that even some of the big boys who spend lots of advertising money have found themselves duped, a case of the biter bit. I wonder how many are coming to the realization that bots generate many of the ‘hits’ they are paying for?
king1394
You make an interesting point. Some will be not just surprised but pleased to see the items they are interested in pop up unexpectedly.
Because you are aware of what is happening you make rational assessments of what is being promoted. Some of the less astute, those that are easily persuaded, may not realize how they are being manoeuvred into decisions that suit the vendor, rather than the consumer.
diannaart
After writing this piece, I was made aware that DuckDuckGo protects searchers’ privacy, does not profile its users, and deliberately shows all users the same search results for a given search term. I must try it.
The picture you paint of the IT coder beavering away creating algorithms that reflect his/her biases rather than the needs of consumers, is what alarms us.
May I clarify, I am not against AI. However I am technically savvy enough to be wary of developing AI’s with too narrow parameters or, even, too wide. It is going to be interesting…
Like Miriam I use free OS (Linux) as well as DDG.
At the other end of the scale you have Morrison purposefully preventing the ABS from using its standard algorithms to smooth the Same Sex marriage survey.
As I only use free websites, I’d rather ads, than the alternative of a million pay per view sites and all the costs and annoyances that goes with that.
Once AI matures more, I would expect a decline in average intelligence over time. If you don’t use it – you lose it. Nonetheless this would not stop me from getting an AI chip in my head.
Algorithms could do my present job – but as my job mostly entails resolving things that occur due to programming limitations as it costs too much to do the advanced programming, I have nothing to fear from AI in that regard.
I also note how pathetic most widespread programs are such as MS Access still are. By now we should be able to talk or use plain English to directly instruct MS Access to get a program written or changed, rather than having to remember the object names and syntaxes.
Extending algorithms to do current human functions will end up undermining humanity via unintended side effects in the longer term. When you do not need cheap labour as robots can do it there will be big consequences – the wealthy will control even more and more than now. Eventually there may be no place for the 90%, particularly as the wealthiest are already merely entertaining themselves in status contests with each other, they don’t need a mass market, they can build robots to contest against each other in their Game of AI Drones. It is more likely however that humanity will have imploded before this outcome could evolve.
astra5
You are welcome.
In the whole considerable time that I have been using the internet, I can still hear those modem strings singing even now, I have never used my real name nor used it for any purchases.
With a few exceptions, I don’t allow cookies and my IP address changes regularly.
When I am concerned about privacy, I use this search engine.
ixquick.com
“Ixquick also provides the stand-alone proxy service”
https://en.wikipedia.org/wiki/Ixquick
https://support.ixquick.com/index.php?/Knowledgebase/List
“Note: If you do not wish to receive Google results, you can visit Ixquick.eu, which returns search results from other search engines.”
I’m quite active on Reddit and I’ve been following the Google censorship discussion for a while. I’ve noticed that a lot of ‘examples’ of Google censorship are really cherry picked. Just try for yourself and see how much of this censorship discussion is true, I noticed it isn’t as bad as people think.
I use Startpage.com (which is Google without the tracking) but I use Duckduckgo to check for censorship (Bing without tracking).
diannaart, jimhaz, Harquebus and Huppie
May I thank you all for your comments, additional information and helpful links. I’ve learned a lot.
DuckDuckGo as described on Wikipedia:
jimhaz, I’ve paid for my website for decades. It lets me tune my site so that it views quickly and easily on any webrowser, on the oldest, slowest computer, even on dialup and ensures my visitors don’t have to put up with advertisements or being tracked by marketers. If you use “free” website you give up all that.
jimhaz, as for your glib suggestion that AI will make us all more stupid, you couldn’t be more wrong. There is every indication that computers have made most of us smarter. (Some people will derive minimal benefit, but that’s always been so — there will always be some who are less smart, because their talents lies elsewhere.) Because of computers I can explore fractal geometry. The way computers let us manipulate text means I can more easily write stories and those stories are better than they would have been if I was limited to pencil and paper. Computers have allowed vast numbers of people to become video creators, with whole new forms springing up as a result. Making interactive fiction was extremely difficult without computers, but now that is a vast and flourishing artform (see for example https://www.ifarchive.org/ and https://renpy.org/ ). Because of computers I can build 3D virtual worlds. A couple of days ago I was converting some computer animations I made… gosh, about 30 years ago and needed to recalculate frame rates. I’ve always been terrible at arithmetic since a primary school teacher used to routinely terrify me for my mistakes. With the computer I was able to create a quick one-line program that rippled through all the combinations of values to print up a conversion table for me. So suddenly I, who have no arithmetic ability can work out halfway complicated stuff. Even though I can’t do arithmetic I love maths, so computers have opened up a whole intellectual realm to me that would have been denied me in earlier times. AI will have the same effect. They’re already having great beneficial effects as doctors checking scan images for cancers miss some that are picked up by AIs, and AIs miss some that are picked up by humans; together they are much better than either alone.
jimhaz, whatever your job is, I expect that unless you are willing to embrace AI you’ll find it will disappear in the near future. http://www.youtube.com/watch?v=7Pq-S557XQU
jimhaz, things could go very badly with the introduction of AI, with a small number owning almost all the means of production (in fact they already do) and the 99% having to scratch for a living. This is already to a large degree the case, but you and I are part of the 1% who are living the good life. The 99% who are doing badly live in other countries. You are just now feeling worried about a future that doesn’t include you in the top 1%. I think it’s most likely that things will go the other way, with democratisation of knowledge and the benefits of society. We can already see the beginnings of a shift in that direction with things like Wikipedia, Wikiversity, Project Gutenberg, Sci-Hub (http://sci-hub.io/ not the false-flag one set up by a publishing company), Patreon, YouTube, Librivox, Archive.org, Ibiblio.org, https://arxiv.org, the distributed Proofreaders (http://www.pgdp.net), PLOS One, Academia.edu, and countless people sharing content for free online. None of this existed a bare few decades ago.