A Little Artificial Intelligence

Artificial intelligence is a hot topic. Depending on who you listen to, it’s the savior or the destroyer of the human race.

Stephen Hawking is a bit gloomy: “The development of full artificial intelligence could spell the end of the human race. It would take off on its own, and re-design itself at an ever increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete, and would be superseded.”

Larry Page, co-founder of Google, is considerably more sanguine in his appraisal: “Artificial intelligence would be the ultimate version of Google. The ultimate search engine that would understand everything on the web. It would understand exactly what you wanted, and it would give you the right thing.”

So, it will either doom us or save us – but what is it?

Artificial Intelligence (AI) can perhaps best be understood from this entry into Wikipedia: “Colloquially, the term ‘artificial intelligence’ is applied when a machine mimics ‘cognitive’ functions that humans associate with other human minds, such as ‘learning’ and ‘problem solving’.”

At Respondology, we use artificial intelligence (also known as machine learning) to help cull irrelevant tweets from tweets that we want to respond to. We want our system to understand what makes a “good” tweet different from a “bad” tweet. And we want the system to get smarter over time.

Let’s look at two tweets that recently came into our system for a client with a product meant for gym enthusiasts.

“Can’t wait to work out!”

“Just found out my SO doesn’t like bacon. I’m not sure this is going to work out…”

You and I immediately understand that we should respond to the first tweet, but not the second. An offer geared towards people working out should be welcomed, but that same ad, directed at the second tweeter, will be deemed humorous at best, but more likely clueless and annoying.

We use the same technology that Gmail uses to reduce spam (something called Bayes Theorem) to make sure that we fulfill our mission with Reply-Based Advertising: to change advertising from being annoying to being relevant. We think that’s a very good use of AI. And while it might not be the savior of the human race, it just might be the savior of advertising.