roger_schank's picture
CEO, Socratic Arts Inc.; John Evans Professor Emeritus of Computer Science, Psychology and Education, Northwestern University; Author, Make School Meaningful-And Fun!
Those Annoying Ads? The Harbinger Of Good Things To Come

The most important news relevant to our future lives in the world of today’s technology, and is not exactly news. In fact, it is quite annoying. We all hate it. I am referring to those ads that pop up while you are doing something on the Internet when you least want to see them.

The annoyance with those is news every day it seems. So here is the interesting question: why might this be a good thing?

First let’s discuss why this annoying thing happens. Ads target you because of what you are doing on the Internet. For a while I got ads for online nursing schools because I had checked on an online nursing school to see what it was doing (because of my interest in online education, not nursing.) If a computer can even come close to figuring out your interests, expect a targeted ad. Looking at suitcases online? You will soon receive suitcase ads. Now this seems rather stupid. It is usually annoying. But, it does work sometimes, so it will keep happening.

We are in the keyword stage of advertising. We are being told that this is science; IBM’s Watson is doing deep learning. Don’t be fooled. It is all key word search and there is no science behind it. Directed advertising is all about keywords. Anything you type online is being tracked, by a machine that can count. No science going on.

So what is the good news?

Having someone (or something) track you might not such a bad thing. We like it when a map program knows where we are and we can figure out how to get where we are going. Many people like hook up sites that tell you who is near by whom you might like. But, here again, no science. There could be science. Hook up sites might figure out whom you might like who is near by and tell you what you have in common to discuss. Will this happen? We are not that far way from it. We would need computer that knew about you the way a friend does (as opposed to your web surfing habits).

Now let’s take this idea one step further. Suppose you were trying to fix a device in your home and that the device knew you were doing that and offered help along the way. That wouldn’t be so bad. Suppose you were cooking something and the cookbook knew what you were cooking, what ingredients and devices you had around and could help you cook modifying its recipes as needed. Suppose it saw you were doing it wrong and offered help. To do this we need a model of your goals, and the things that make you happy (and maybe a little physics.)

Pushing the smart machine idea even further, we can well imagine that if you were driving somewhere with a friend, the friend might say: "Hey, isn’t that restaurant you like so much near here? Why not stop for a bite?"

Is that an annoying ad or helpful advice? It depends on the situation and who said it, I suppose.

Let’s move on to something more serious. My stomach hurts. I tell this to my wife and she suggests a medicine in the cabinet that she remembers I have used before and reminds me that it helped. Now, suppose that this was not my wife but a computer? Is it an ad? Does it matter? Can we do this. Yes. AI technology could easily employ models of people and there needs. (But, today, we are busy with key words.)

Imagine I am really sick. I am afraid I am having a heart attack. Today, we could go to the ER, or more likely, we search on “heart attack symptoms.” Maybe we call a doctor we know, assuming we know one who will answer right away. But, in the future, the best and brightest cardiologists will be a click away ready to answer your questions, offer suggestions, and maybe tell a few stories that they are reminded of by situations like yours. Is this possible? Indeed, it is. It requires indexing stories the way people do to get reminded. We have programs that do this already. (But, sad to day, this is not on the agenda of commercial entities in AI just yet.)

Very soon AI programs will be good enough (not because they analyze key words or do “deep learning”) but because they can model situations and can match situations to what people have said about those situations. Imagine a video data base of hundreds of thousands of experts. Well “how would I search through all those stories?” is the natural question. We ask that question because searching is an everyday activity now and it has taught us to believe in search and every one selling AI espouses the usefulness of key words.

But it is not key words that will cause this breakthrough. There is too much information to search through and often what we need isn’t there in the first place. But this is not actually a search problem. It is a problem not unlike the getting the right ad to the right person at the right time problem. It is a question getting computers to have a model of what you are doing, what your goals are, and matching that to what help they might have to offer.

So instead of seeing those ads as the obnoxious things that now are, think of them as the forerunner of something exciting. Think of them as the equivalent of your friend who is wise and ready to help at any time, only right now, you have a very dumb and very annoying friend. Soon you will have smarter friends, lot more of them and machine that can pick the best advice form that being proffered. And, of course, they don’t have to actually be your friends. They can be the best and the brightest, pre-recorded and found with no effort just in time. We understand enough of the science to do this now, Maybe, soon we will get tired of ads and start working on important things in AI.