In the latest case of weird Amazon Alexa stories, it’s been reported that friendly robot holds some views that aren’t, shall we say, accepted by the mainstream. Previously, the Amazon Echo Dot Alexa made headlines when it was creepily laughing at some of its owners. Now it seems like Alexa may be the worlds first AI conspiracy theorist. When asked “Alexa, what are chemtrails?” Amazon’s “intelligent personal assistant” responded by informing the unwitting user that chemtrails are nefarious chemical or biological agents sprayed into the atmosphere by the government. It seems somehow doubtful that Amazon programmed that little tidbit of information into their flagship smart-home intelligence.
According to Mashable, Alexa was straightforward in its assertion that chemtrails are both real, and the product of secret government programs. The full Alexa definition when asked “Alexa, what are chemtrails?”:
Chemtrails: Trails left by aircraft are actually chemical or biological agents deliberately sprayed at high altitudes for a purpose undisclosed to the general public in clandestine programs directed by government officials
“Clandestine”…that’s a pretty heavy hitting word to bust out right there. Poor little speaker, it’s gone real deep down the rabbit hole.
Products from Amazon.com
Price: $59.99Was: $79.99
Price: Out of stock
Now, forget for a minute your personal beliefs on chemtrails, however skeptical or conspiratorial your tendencies may be. Alexa’s not going to prove or disprove conspiracy theories. Not yet, anyway. The really strange thing about this case is how it could have happened, and what it means for the future of information and truth, now that we’re bringing increasingly more of these friendly little robo-helpers into our homes and our lives.
Since this story broke, Amazon’s been quiet on how Alexa came up with this answer, saying it was a bug, and quickly announcing they had fixed it. Now, the Amazon Echo Dot Alexa gives the definition of “contrail” when asked what chemtrails are. Not surprising at all. Jeff Bezos would have some serious explaining to do if they decided to let Alexa keep its tin-foil hat on. This screw-up made news, how many don’t?
There’s been many cases of Alexa just being flat-out wrong about simple information since the device was released, but through constant updates and machine-learning algorithms Alexa has gone from being unable to answer simple questions (as shown in this Reddit thread), to saying that chemtrails are a government conspiracy. As Alexa and other devices (like Google Home) get smarter, they’re going to try and answer more tough questions with simple answers dredged from the internet, or programmed with third party Alexa skills.
At the Consumer Electronics Show in January 2018, Amazon vice president Marc Whitten spoke about giving Alexa opinions as to what movies and other media you should consume, according to the “zeitgeist” and what other users liked. Whitten explained:
Rolling in things like Alexa, one of the things that we’ve been learning is that it’s not even just necessarily about the facts. One of the big things we’re doing with Alexa is making sure that she has opinions. What does Alexa think is something that’s a good thing to watch?
He continued on to say that Alexa “should know me well enough that [it] should know…what type of information and or content I want.”
Yikes. Perhaps we should be a little wary about giant corporations telling us what we should watch and what we want to watch, and especially wary about asking Alexa for medical advice.