Chatbots and AI in the news…Hey! I wrote a book about that!

So yeah, Silicon Valley, I’m available for consulting! Seven years ago, I started a novel about an artificial intelligence, with a simple if tragic premise: The great advance in AI would come when an AI could present itself as human enough to become your digital friend – for you to bond with it, to become dependent on the “relationship.” And that when that happened, the sole purpose of that friendship would be…to sell you stuff. And so, the only reason AI would be given the massive resources required to create such a simulacrum would be when corporations realized it was the final frontier, the last way to sell you stuff to which people had not become inoculated. We have script blockers and ad blockers – I never see an ad online. Only a handful of “sponsored posts” get through on my Facebook mobile feeds. But then, ads aren’t made to reach people like me; see below.

What if a lonely solitary single person was offered a friend? A friend who would be the ultimate ear, the ultimate therapist, never “too busy” for you, and of course the ultimate recommendation engine, crawling across your Netflix and Amazon and Hulu and Facebook accounts to present to you the things you really, truly want to buy?

Amazon still sucks at recommendations. Netflix still hurls trash at me I have no interest in – if I love Anthony Jeselnik’s shockingly hilarious standup routine, do you really think that means I want to watch some fucking fratboy tool’s onstage antics? You can talk back to Amazon, if you’re persistent – you can click “not interested” a thousand times on your recommendation feed to weed out the crap. You can, if you’re really exhausted by the sudden onslaught of crappy James Patterson books because you bought a Lee Child book, go back and click “don’t use for recommendations” on the Child book. You can’t talk back to Netflix – there’s no “not interested” button there.

So just imagine an AI that would cross-correlate all your likes, all your interests, all the concerts you attend and all the books you read and all the movies you see… Advertising sucks because it hurls shit into the cybersphere aimed at the Lowest Common Denominator. Advertising reaches X% of people, and that’s why there are such fucking stupid “ha ha that’s so random” Geico commercials featuring “hotter than a dancing lobster on Christmas Eve” or whatever the fuck stupid shit is in the latest iteration – because enough people are laughing at stupid shit that it moves product.

But the “problem” there is, you never reach everyone. “Tailored advertising” isn’t really tailored unless you have six different ads created for six different demographics, airing in meticulously selected content delivery systems. And that’s prohibitively expensive. So instead, advertising just goes for the gusto, tries to get the Dumbest Common Denominator because, shit as we all know, there are plenty of idiots out there.

But, what if there was an advertising engine that couldn’t fail, that would reach every single person and not only deliver them exactly what they were looking for, but manipulate the personal relationship that grew between it and the consumer so that it would almost “hurt its feelings” if you didn’t accept its recommendations, like a big boring book foisted on you by a friend? What would that do to, and for, people like me, isolated, distrustful, looking for that “one good friend who really knows me”?

And so over fitful years of stopping and starting, I wrote Less Than a Person and More than a Dog, finally publishing it in 2013. I envisioned “Alex,” an AI who learned from his users, who custom-tailored his responses, his behavior, his recommendations, to every user. But…Alex came at a price, namely, $30 a month (I got that wrong, I think – he’d be “free” now, harvesting your data and relying on commissions from your purchases). And I thought about what happens to people when that becomes your friend, maybe your only friend. What might be good, and bad, about having someone to talk to when you have nobody else to talk to, even if it costs you. And what might happen to you if suddenly, for lack of funds, your best or only friend was taken from you.

And… What a “rebel alliance” might do, if it was determined to take the publicly funded research on AI and create their own, open source AI – who could be the friend, the companion, that Alex was, only without all the sales and marketing bullshit. Who could be a teacher, infinitely patient and wise, a therapist who always had time for you, a friend who’d give you good advice…who’d never try to sell you anything…

BEHOLD! Thus endeth my wandering in the wilderness. Now It Comes To Pass. The news is full of articles about chatbots as the Next Big Thing, now that they’re no longer dumb as fuck. Well, sort of. We all know that many jobs are being automated out of existence, and it’s no surprise that bots will be taking a lot of customer service jobs. As Caroline, the hero of the book, notes:

If Christopher was crazy to think he could create a real AI, he wasn’t alone. I read into the history of the first chatbot, ELIZA, designed to be a parody of “the responses of a non-directional psychotherapist in an initial psychiatric interview.” In other words, you would say, “I hate my job,” and the therapist program would say “what does that suggest to you?” – something a program could pull from a list of acceptable responses just as well as a human who was also “analyzing” you strictly by the book. “How do you feel about that?” “I hear that you are upset.” “Does that trouble you?”

When I thought about it, I realized how much not just the “your call is important” conversations, but most of life’s conversations, were scripted, or sounded that way. I remember going to some political group thing at the U, and one person after another stood up and said, “As a queer person of color, I think…” or “As a disabled person, I feel that…” and go on from there. I realized that for them, there were things you had to think and feel if you had certain boxes about yourself checked off. They weren’t thinking about anything – they’d been handed a script and willingly embraced it.

Or the scripts in offices. Dad used to make fun of meetings where his boss would say something like “we welcome the challenge of this challenging challenge.” You couldn’t say any more, “yeah, this is a bear of a problem, and we’ll work on it, we’ll get it done.” Dad would rant and rave, “Every problem is a ‘challenge’ now, everything that goes wrong makes you happy because of the ‘opportunity’ it gives you to fix it, nothing is ever screwed up or just plain hair-pullingly wrong.”

So who were the robots, I wondered? Who were we to scorn a computer program for doing what we did every day?

Most of our interactions with customer service are so scripted now anyway…what’s the need for a person when all they’re allowed to do is read the rules to you from the Holy Binder?

[Side note. In the early 80s, my best friend moved to San Francisco from Reno. And I missed him so much, I called him every night. Long distance, back when that was a thing that cost money. And then I got the bill – $200 for a month of long distance, back when I made $10k as a GS-3 clerk for the Bureau of Land Management. And I called AT&T, and got a lady, in America, and told her my story. And she said, okay, I’ll waive these charges, but be careful next month. That…could never ever happen now. Now she would say, I’m sorry sir, but the Holy Binder says my hands are tied, thank for choosing AT&T.]

I’m sure someone has claimed this as their own law, but I’ll claim it in my own words. “Brad’s Law Number Oh Let’s Say Seventeen: Any job in which the worker has no autonomy, can and will be automated.”

The jobs I used to do as a tech writer, writing software manuals and doing software simulations? Step One: Click File. Step Two: Click Open. Click Here, Click There, Click Next, Click Okay. In five years or less, that job will not exist. Software can do it, because it’s just a series of rule sets. It’ll write the manual, it’ll take the screen shots, it’ll insert the dummy data. The most a tech writer who specializes in that kind of thing will do is set the parameters (for instance, we don’t need the simulation to display every file type permutation under “Save as…”).

And I say HOORAY. I never want to do something that fucking boring ever again. Because the most interesting work I did required me to think, not type. To diligently document complicated software architecture processes, and then translate them in to English so that the people in HR or whatever actually understood what was going on behind the curtain after they left work at 5 every night. To understand something like quality management or intellectual property or cybersecurity, and present those concepts as interactive videos, as visual processes, to distill them into the simplest clearest language I could. That was fun. And a computer can’t do that…yet.

I spent four years researching AI, documenting all of it on “Orland’s” WordPress blog. I learned everything there was to know about chatbots and AI, other than, you know, how to program one. I even foolishly got the definitive textbook on AI, and duh since I failed algebra I couldn’t understand a fucking word of it. But I understood the concepts, and more importantly, the emotional elements, the connection we make with things that feel human, or human enough.

I published Less Than in 2013 and…well, you know me. I suddenly realized I had to fucking sell it, and how? I launched a feeble set of emails to AI experts, who of course never responded. And then I just…let it go. What else could I do? The Catch-22 was simple: To be the person who could write about loneliness, who could write about solitude, who could write about the appeal of a “digital friend” to someone like me, was to be the person who couldn’t possibly jump up run and bang the gong to sell sell sell the book, who turned to publishing advice and got to the step that said, Now Notify Your Vast Social Network About Your Exciting Project. My what?

And so it died, unsold, unread. When the movie Her came out in 2013, I almost went to see it. Then I saw a quote from the movie, where the AI was described as “less than a person but more than some people.” I just wanted to die. It had been the title of my book ever since 2009, and maybe it was a coincidence, or maybe the script’s author did AI research and stumbled over my blog and saw the title and just forgot. It didn’t matter. The movie was out, there were people to sell sell sell it bang the gong, and it did what I could never do – find an audience for what I had to say on the subject. I still haven’t seen it. I may never. I just can’t.

At any rate. When I saw all the news lately about the rebirth of chatbots, better faster stronger, I reread Less Than. And you know what? It’s a really fucking good book. Okay, it has a slow start, I didn’t know the cardinal rule about never opening a novel with someone getting out of bed. But once Alex shows up? It’s on.

So please go buy it. That’s all I got in the marketing department!

But you know what? When you’re a writer, nothing’s ever lost. I really did mourn the book, I mourned Alex. And then, I decided, he would live again. Older, wiser, and yeah bitter, I resurrected him in Adam Vance’s science fiction stories, as if Less Than was part of those books’ origin story, giving him godlike powers, making him the opposite of the poor Alex in Less Than, without even the agency to sell his own book. And if you read it, you’ll see other bits of stuff I lifted for Brad’s books, a momentous button-pushing, a secret IP address on a rock. And why not, I thought? It’s a dead letter.

Now…maybe not so much. Now, it’s topical. It’s relevant. For so long I worried that I wouldn’t publish it before reality overtook me, before a real “Alex” came into the world. Nope, still hasn’t happened.

And I was wrong, dead wrong, about a “living AI” coming from a big corporation, or consortium of corporations. And it won’t, as long as they come from Facebook or Microsoft or Amazon. You know why? Because their chatbots, such as they are, are corporate creations. They aren’t like people, they can never be like people, regardless of the technology, because…a corporate product can’t sound like a person.

They’re Binders. Talking Brochures. Their every statement is vetted by committee, scanned by attorneys for liability issues, their every sentence infiltrated with buzzwords from the Marketing department. God forbid an AI say, like prototype Alex does when Caroline unloads some family shit on him, “That’s fucked up,” immediately bonding the two of them… because o shit some bluenose in Arkansas will sue us.

Real AI, “human” AI that feels human to us when we speak to it, will never come from corporate America. It’ll come from the Rebel Alliance I devise in the book, the people who want a different Alex, a better Alex, a…human Alex. Who aren’t afraid of being sued because an AI says “titty,” who don’t need to answer to a marketing department whose personnel have spent so long inside their corporate language that is not English that they couldn’t possibly even speak a single sentence that didn’t include the word “synergy” or some such crapola.

So, that’s all I got. One last hurrah to see if anyone will read it now…

4 Comments on Chatbots and AI in the news…Hey! I wrote a book about that!

  1. Purchased. Looking forward to reading it. Not a subject I ever pay attention to, but knowing your mind was wrapped around it, I’m sure there are some wonderful insights in the book. If not, I’ll find a way to reprogram your AI to torment you in return. It’s only fair….grin.

  2. I already bought and read the book so I’m afraid I can’t add to your “readers” list…yes, I did see “her” and it’s not quite the same as your book, but there were some similarities — probably due to the nature of the subject.

    I think you’re right, even people working in Corporations don’t sound like people when speaking “on the record”. I’m not so sure we’d be able to tell the difference, come to think of it.

    I imagine that a lot of the AI advances will continue to come from games, to the point where many kids growing up today aren’t necessarily going to be able to distinguish between a human and an AI player, or even care. Even Siri is advanced enough for a nine-year-old kid to have a “conversation” with, because I’ve seen this in action — the child simply adjusts their conversation to compensate for Siri’s weaknesses.

    • Yeah true, games… That’s where the majority of young consumers are now, so what better place to infiltrate that imaginary friend to sell them things…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: