Bots and social media engagement

Opinion
Nov 16, 20103 mins

As people in the online world become increasingly connected, the reasons to engage with social media expand with remarkable speed. Just consider that only a couple of years ago the leading edge of the Twittersphere was the Los Angeles-based mobile Kogi Korean BBQ which was using Twitter to drive sales … now, there are scores of similar players all over America using the same strategy.

An interesting side note is that the L.A. mobile food world has, in fact, expanded phenomenally and there’s even an aggregation service that tracks most of the vendors (though it recently seemed to be in FAIL mode).

Anyway, what got me thinking (again) about the opportunities in the Twittersphere was a clever Twitter chatbot (an automated Twitter service) that aims to addresses a very hot and contentious issue: Climate change.

Whether you agree with the concept that the Earth is undergoing significant climate change or not, if you are on either side of the argument you most likely want to support and or promote your viewpoint. On the “it’s happening and you’d better get real” side of the discussion is the Twitter chatbot @AI_AGW.

Created by a software developer, Nigel Leck, who apparently “got tired of arguing with anti-science crackpots on Twitter.”, @AI_AGW searches Twitter for key phrases that indicate another Twitter user is espousing anti-climate change sentiment and attempts to engage them with tweets that put forward a counter-argument supported by scientific evidence.

It appears that Leck’s chatbot also tries to drive discussion on the topic by posting tweets that entice other Twitter users to respond, a very cunning way to increase the visibility of the “meme”.

While @AI_AGW does a pretty good job, what it lacks is a certain sophistication in understanding the intentions of the people who interact with it … for example, its ability to understand sarcasm and irony is non-existent so it will challenge people making comments such as “when a record heat wave hit L.A. last summer, causing innumerable tweets of the form ‘It’s 113 degrees outside – good thing global warming’s a myth!

What’s interesting is that Lack’s chatbot is illustrative of an important trend: The shaping of public opinion by automation. Leck’s bot could, quite obviously, do well with a lot more support for linguistic analysis to improve its relevancy but even so, its impact on opinion (given the low-touch, high-speed nature of the Twittersphere) should not be underestimated.

A similar strategy to promote a less intellectual argument such as “Pepsi is better than Coke” or “vote for proposition X” could have a significant impact if the bots have a more believable appearance (in other words, as if they were human). Moreover, if there are many of them all performing slightly different styles of engagement, then it appears that many “people” are driving a viewpoint and the possibility of a meme being going “viral” becomes much greater.

This, my friends, is the way of the future or, maybe, the way of now … which means when it comes to the issue of how you form opinions, do you really know who you’re engaging with on Twitter or any other social network?