4 Words: No Bots For Now. Also, will be interesting to see just how limiting using only Apple’s owned products (Apple Pay, Maps, etc.) will be on the success of this. Could this be interpreted as a ‘mea culpa’ by Apple of how they defined an app ecosystem with too much friction to download and unintuitive discoverability?
An interesting look at the relationship between data and automation. As the author puts it, “Data science is an input and the output is yielded in the form of automation. Machines automate, but humans provide the necessary input to get the desired output.” Will this lead to partnerships between chatbot companies and big data analysis outfits? Chatbots = something to actually *do* with the data? Hmm?
We like this article because it actually shares some stats and success metrics (in this case 20% increase of conversions via bot vs. web form). But exposes some challenges the ad-to-bot flow on Messenger still has. Especially on Facebook Lite and Messenger Lite. Would love to hear how you’ve see this perform!
At Reply, we’re pretty vocal about our thoughts that the objective of a bot should be to provide a service, not to mimic a human being. That said, we think fascinating peek under the hood at Facebook Artificial Intelligence Research (FAIR) is an example of where those lines get blurred. Because so often, the need to speak to a live human is do to the desire to negotiate, to get a better deal. So effectively, the work these folks are doing while it is mimicking human behavior, the objective remains: to provide a better service. Things that are a bit chilling? There were cases where agents initially feigned interest in a valueless item, only to later “compromise” by conceding it — an effective negotiating tactic that people use regularly. This behavior was not programmed by the researchers but was discovered by the bot as a method for trying to achieve its goals.