I know a lot of people already know about this, but I thought it was an interesting topic to put up for discussion.
Even if you did know that machines watch you surf the web, by the time you found out it was already too late. The technique called machine learning “affects the kind of Internet advertising you see on Amazon.com, but most people don’t realize that is the underlying technology” (livescience). The ads we see online are often put up by machine learning systems to “learn” how people use the Internet. By comparing your activity with other users, the “agent” targets what ads you are most likely to respond to. Then, every time you click on an add, the host site (usually Google or Facebook) charges the advertiser.
So how do machines learn? They start out at the level of a child and have to learn more to progress further in “intelligence.” For example, the article mentions that “like school children learning arithmetic from a teacher, some agents can use simple examples and feedback to learn how to approach more complicated information.” Other agents rely on experience: using the responses from their past “decisions” to deal with similar situations. What is most important to notice is that these machines figure out the patterns in the data they receive. The downside to teaching a machine to learn is that they are doing so based on a model of human learning. This causes some issues because we don’t fully know how the human brain works. So for now, these machine learning programs can only process a limited amount of information.
So, what does this mean for us? It probably means that advertisers are going to get better and better at reading the individual’s wants, so that everyone is constantly being barraged online by things they really want. It took decades for advertisers to hone their methods to personalize ads, but machines are already able to do it faster and more efficiently. When the technology is perfected, consumers’ only option will be to “just say no.” Watching a commercial for something is awful, boring, and frustrating. Ads placed on websites are hardly notices because you’re focusing on whatever else you’re doing. If you do click on those, you certainly don’t feel like you’re saying “yes” to purchasing something, but you totally are.
I’m sure other (and much creepier) implications can be drawn from this too. It presents a great “Big Brother” opportunity for everyone from the ad agencies to the government being able to monitor what has become a pretty crucial part of our consumer culture: what we want. Professionals can build complete personality profiles based on what ads you respond to, and who knows who they would be willing to sell those to? Anyone?
What worries me the most about this innovation in technology and advertising is that children are easy to be influenced by this. I mean, we all are, but it definitely isn't fair when it comes to child advertising. There are already many issues about advertising to children in general. However, this raises the issue of digging into the minds of children, creating ads specifically for them, and making them want a product/service. Unlike adults, children haven't developed a strong identity and have definitely not developed the will power to "just say no" either. There needs to be a line of ethics with this issue.
ReplyDeleteShould this information only be sold to advertisers of specific products? If so, what products? To what products should it be restricted to?
Addressing your concerns about advertising: this has been Google and Facebook's model from the start. These entities justify their use of these advertising practices by saying they are giving us "free" services and these ads are how we pay for it. This is true, but their must be more oversight on what they are doing with our information. Creating a sense of indebtedness in us allows them more freedom with our information.
ReplyDeleteThis raises another question, however. Would you be willing to pay a monthly fee in exchange for no advertisements?
Personally, I have to question the 'accuracy' of targeted web advertising, at least in its present stage of development. Not all of our web browsing habits are consumption driven... I for one will more often use Google to look up information about a person, place, or event than to look up products and/or services. What this results in are ads on my browser that don't really correlate with things I would actually buy. Amazon is a different story; since their ad machine grabs data from my actual shopping history, the chances of it hitting me with a link I would find appealing are considerably higher. I am actually grateful for this, as it has a way of introducing me to interesting items -- usually books, CDs, and DVDs -- that I probably wouldn't have discovered otherwise. I doubt there's any kind of technical flaw in Google's approach... it's just that people's web browsing habits are often about as predictable as people themselves, which is to say not very.
ReplyDelete