Read an Excerpt

CHAPTER 1

More Nude Scenes

Man, in the large, is a mass urged on by a force.

— Tesla

Algorithms are amazing, vicious, poetic, ruthless, stupid and genius. As the DNA of modern computers, they are elegantly simple, exceedingly complex and central to the apps on your phone, artificial intelligence and all the software or devices offering suggestions, filtering, sending notifications and/or making complex decisions for you. In short, an algorithm has directed your life today. It is only natural, therefore, that a seminal book on software studies, which goes by the same title, has "algorithm" as its first entry. These subtle lines of code dancing on the programmer's page are the past, present and future of computation, and they are the one thing computer scientists agree upon — "Without the algorithm ... there would be no computing." Existing in mathematical models and in abstraction, either in our brains or on paper, an algorithm comes alive when it is embodied in computer hardware; they have to be materialized to have an effect in this world. Just as a Hollywood script needs actors for it to become a play, in order for an algorithm to flex its muscle it needs data (inputs) and, as we are all now aware, data can be anything.

According to Thomas Cormen, author of Algorithms Unlocked, "Computer algorithms solve computational problems. We want two things from a computer algorithm: given an input to a problem, it should always produce a correct solution to the problem, and it should use computational resources efficiently while doing so." In other words, algorithms are a formula or procedure for telling computers what to do, and need to be consistent and efficient in their use of data when providing a solution. Computers are algorithm machines. Consider the rest of this chapter as a type of algorithm — except the "input" is the algorithm itself, being fed into culture, and the "output" being its effect on our cultural health.

"More Nude Scenes"

Algorithms help Google Maps get you home when there is an accident ahead and you need to get off the highway. Do you go east off the highway? Do you go west? Now that you have gone east off the highway, what street to take next? The first street on the right has a lot of lights down the road, but the third left is a meandering single lane prone to build-up during rush hour. Is it rush hour? What is the weather? What happened last week? What is the trend for this route — has congestion been increasing or decreasing over the past month, week or day? What is the data being tracked from other drivers' phones saying? All this information (inputs) is fed into algorithms to get you home the quickest (an output in the form of a route). It is a simple process, but human minds cannot process a fraction of the data as quickly as your GPS, which is why computers started beating human players in chess (in the late 1980s) the moment they had more speed and processing power. Many experts thought technological progress would claim its advancements on the back of processing speed, Ray Kurzweil, now Director of Engineering at Google, being a famous proponent of linking processing speed and capability, but algorithms may be the true revolution. In one report, drafted by the US President's Council of Advisors on Science and Technology and titled, "Report to the President and Congress: Designing a Digital Future: Federally Funded R&D in Networking and IT," the following observation is made — "Even more remarkable — and even less widely understood — is that in many areas, performance gains due to improvements in algorithms have vastly exceeded even the dramatic performance gains due to increased processor speed."

Output of picture gallery one: our decision-making capabilities regarding geographical orientation and awareness are being eroded.

Algorithms are allowing political parties to cater their marketing material specifically to your neighbors. For instance, did your neighbor lose their wife recently? Well, big data can predict what type of political narrative will most likely appeal to your grieving neighbor, perhaps a narrative of empathy and compassion with tidbits of injustice. Social profiles of us are already available for purchase on the Web and political parties can purchase those profiles, feed that data into their systems and spit out brochures, robo calls and emails "perfectly" tailored to your existence.

Output of picture gallery two: the political intelligence needed to understand issues in communities has given way to data tracking and automated analysis, thereby rendering some parts of the "boots on the ground" political process useless. The only logical result of this trend is voter distrust, cynicism and anger, which will change the face of politics in each sector where this is implemented.

Algorithms are not only making inroads into politics on a marketing level, but on a policy level as well. Valentin Kassarnig has managed to develop a program able to write political speeches that are, apparently, difficult to determine if an underpaid staffer penned it or a computer wrote it. For instance, it produced: "Mr. Speaker, for years, honest but unfortunate consumers have had the ability to plead their case to come under bankruptcy." How did Kassarnig do it? By using an algorithm to scan and analyze 4,000 political speeches, find patterns, and then produce results implementing those patterns. Once the algorithm knows what is effective and going on in speeches already written by human minds — word length, affective tone, frequency of verbs used, etc. — it can clone that pattern and output an analogous speech, and as you can see it does a fairly good job. Users of Gmail now have the option, called Smart Reply, of semi-automated email responses, wherein a program analyzes an incoming email and then offers a few responses, replete with the right tone and all.

Output of picture gallery three: the skill set to construct public intellectual products such as political speech will render an entire class of political intellectuals useless. While this is a small example, when applied elsewhere the ability to construct meaningful sentences for public consumption will become a useless skill, along with the thought process needed to create such speech.

When algorithms and pattern/association matching processes are brought together you get apps making decisions for us, apps based on algorithms that, when combined with other soft and hardware, gives rise to something feeling like artificial intelligence. I am referring to the culture of the playlist "Recommendation" or "Because you watched ..." A more complex and recent phenomenon regarding content creation, the entertainment industry already uses algorithms to predict the box office success of a film, an algorithm using characters, plot and actions as input "data points." Epagogix, the developer of this "smart" script reader, describes their product in the following manner:

Advanced Artificial Intelligence in combination with proprietary expert process enables Epagogix to provide studios, independent producers and investors with early analysis and forecasts of the Box Office potential of a script. Clients then make evidenced decisions about whether or not to spend their scarce capital, adjust budgets, or to increase the Box Office value of the property.

In other words, their algorithms can tell a producer that their film starring such and such an actor will only pull in 20 million as opposed to the 35 million the studio thought it would ... before the filming even begins. With the help of Epagogix the production team can decide to get a more attractive, lead female character (a bigger net gross), add more nude scenes, or keep things as they are and pay the actress less, since they are confident the film will underperform and their ROI outlook is now revised.

Output of picture gallery four: our ability to cultivate a personal aesthetic is becoming unnecessary. Moreover, the creative individuals who read, edit and review such materials will soon become useless, as the act of reviewing and editing will become automated. All you have to do is combine picture gallery three and four.

Algorithms are beginning to determine what books get published from public domain. January 1 of each year is called "public domain day" for many publishers, as this is the day the copyright sticker on books written decades ago is torn off. One researcher has found out a way to mine the views, clicks and article lengths of Wiki pages, in combination with data from a database of books already in the public domain, and, when put together and processed accordingly, a ranking can be generated. Publishers can then use this ranking to decide on which books to publish.

Output of picture gallery five: the intellectual industry of publishing that looks to great works of art and cultural trends to publish books for public consumption is becoming obsolete, with effects that will trickle down not only to a decreased appreciation in books and cultural teachers, but also in education programs that hone this skill.

We now have a coat that hugs its wearer — from behind — when it detects said individual needs a hug and is openly termed a substitute girlfriend; to be fair, it is only a prototype at this point. The wearer connects to headphones and along with a hug, there are prerecorded emotional voice bits such as, "I'm sorry, were you waiting?" "Watch your back!" "Guess who?" "Blind side!" The fact that the coat hugs from behind, so as to catch the wearer by surprise, like a typical girlfriend would do (?), can, with little stretch of the mind, be best termed an existential cough drop — akin to professional cuddlers or services that send girlfriend'esque text messages to men who feel lonely — a bit of false humanity to stave off modern anonyme. And, given that the voice is from Yu Shimotsuki, "known for work in adult-themed video games," there is a seductive porn edge to it all.

Output of picture gallery six: products are slowly chipping away at social initiative or the motivations thereof, situating themselves between our biological signals of, for example, loneliness and their solution, effectively, nay potentially, rendering ourselves useless in the process of knowing our emotional states and modifying that emotional state.

Lest we get the impression that algorithms are trivial entities, they are being deployed by over a dozen young startups to fight cancer and ultimately could save thousands of lives in the years to come. IBM, BERG Health and CureMetrix, to name a few, are using AI algorithms to interpret body scans for cancerous growths, to detect minute changes in temperature from a woman's breast tissue so as to raise an early warning sign of breast cancer, and to help companies, such as Globavir Biosciences, reduce the time it takes to develop new drugs.

Output of picture gallery seven: the skills of professional medical researchers are being enhanced, not replaced, by AI algorithms; however, the utility of the researcher will diminish the more sophisticated our software becomes.

Related to the hugging coat, there is the T. Jacket, about $549 USD at the time of this writing, which is a jacket worn by kids allowing parents, via a network of air pockets and sensors, and, of course, an app, to hug their kids remotely. Johnny gets upset and smacks Cassandra at school. Johnny's pulse and body waves indicate he is distressed. The jacket, set by the parents to activate at certain thresholds, inflates and hacks the body-love system, that is, intervenes between biological response and stimulus, and, ideally, calms the child. The invention is backed by research about touch therapy for certain individuals, as well as by certain forms of mental conditions, where pressure on the body significantly helps the individual feel secure. Also, software is beginning to decipher the cries of an infant. Is your 2-month-old baby hungry or distressed? Who the hell knows most of the time!? They are two months old. No one knows. Yet, the sound waves and types of cries are treated as inputs into software, and the output is an interpretation of your child's fully inarticulate yelps and gestures.

As you can see, we are well beyond functional algorithms: they live between us and our instincts, between parent and child. How far of a stretch is it to imagine that, in the future, when the software detects a hungry infant, a machine automatically makes a bottle, and, for the adventurous types, a robo-maid delivers the bottle to the child. Do we applaud this move, since now mothers and fathers, so pressed for time, can focus on something other than guessing why their infant is crying and holding them or swaddling them or rocking them to no avail? Or is this a Trojan horse, with the inevitable result of eroding basic human bonds at one of the infant's most vulnerable times in their lives?

Output of picture gallery eight: software is being situated at intimate levels at stages of human bonding and can render the parent's instinct useless, or at least degrade their confidence in listening to their children. While this is a small example, the applications for this technology in all parts of behavioral and speech recognition are tremendous.

There are belts to help with overeating by analyzing our biometrics and generating a real-time diet, and Microsoft has developed a bra to help women make better food decisions as they near the fridge seeking a treat. Another company, this time Japanese lingerie brand Ravijour, has created a bra set to unhook only when its sensors indicate that, indeed, the wearer is in love. Fundawear, a product developed by Durex, is allowing users to play digitally with their lover's body — yes, in those parts — by an app. INTIMACY, a line of avant-garde fashion clothing coming out of Studio Roosegaarde, can make its fabrics transparent by reading the data on the body. So, presumably, based on the signals it is receiving from the wearer's body, your clothes turn transparent. I can only imagine the hacking potential here on Friday night at the bar, surely to become another form of digital sexual assault.

On some software, it is possible for women to log their menstrual cycles, to help them better define the times when they are most fertile. In other words, an app can now tell you when to have sex. Another company has patented an electronic orgasm delivery devise that is apparently only for medical purposes, such as for women with "orgasmic dysfunction." Able to deliver an orgasm with the push of a button, we should believe this machine will only be used by the dysfunctional with as much credence as we think the Internet is only used as a tool for enlightenment.

Output of picture gallery nine: while seemingly trivial, the presumptuousness of software to hack our ability to read our bodies at the level of "love" is aggressive and somewhat humorous, like a gag gift from the mall, but the use of diet software will cause a decrease in our confidence to read and understand ourselves, make informed decisions about how we consume food and for what reasons. Akin to the jacket that automatically hugs its wearer, the orgasm delivery device is further grease on the wheels of social alienation and personalized pleasure delivery.

Even the highly nuanced procedures of legal analysis and documentation, it appears, is under the threat of automation. While a trove of headlines lamenting lawyers will soon be replaced by software able to sift through, find and analyze the relevant cases for determining a defense, white paper or procedural strategy, movements have already been made in this direction. The discipline of psychotherapist evaluation, often labor intensive, could give way to programs listening to patient-therapist conversations, transcribing the conversation to text, analyzing the text for "counselor empathy" and then providing a rating. Apparently, early efforts of software to detect counselor empathy against traditional human interpretations have proved close. Naturally, it is only a matter of time before the software is offering suggestions to the therapist to make them more empathic. Moreover, once the metrics from the social situation or whatever are made available, real-time, to the software, it will further be able to predict the patient's mood, then send tips or directives to the therapist in real time for implementation in a session. Say a patient is a good liar and, though seemingly aggressive, is actually falling in love with the therapist (a common problem), then cautionary measures can be taken immediately; perhaps the algorithm made this determination based on body heat, body language, doublespeak and eye contact, wherein it might have taken the therapist a few sessions to realize. In a boost to Google's self-driving car ambitions — and the desire to humanize software — in 2016 the US National Highway Traffic Safety Administration said that it considers the "self-driving system" to be a "driver." Analogously, software will soon be considered a "therapist."

---

Excerpted from "Society Elsewhere"
by .
Copyright © 2017 Francis Sanzaro.
Excerpted by permission of John Hunt Publishing Ltd..
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

---