‘M’ Assistant – a New Territory For Advertising
‘M’ is an Artificial Intelligent virtual assistant inside Messenger which aims to be helpful to its users when appropriate. M Assistant analyzes their normal conversation on Messenger and pops up to suggest one of the features within the app such as create a plan, make a call, leave a voice message, send a sticker. It was launched in April 2017 in the USA and it keeps adding new features and expanding to new countries. At the time of writing the AI Assistant is only available in English and Spanish. M Assistant brings a number of controversial issues concerning harvesting Big Data, surveillance and privacy. It provokes questions like how is ‘M’ Assistant going to change the way we search for information and ultimately, how is this going to affect advertising?
Messenger was called ‘the new social living room for the world’ by its Head- David Marcus. If Messenger is the living room for the world, one can assume that Workplace is the office, Instagram is the cupboard with the photo albums, Memorialized accounts are the cemetery. Apparently what Facebook aims is to recreate our whole world. Or with its bots to bring the whole world to our social living room, so that we wouldn’t need to go anywhere else.
The beta version
In the summer of 2015 Facebook let a bot enter the living rooms of a number of ‘invited only’ users for its research and development of project ‘M’. Unlike Google Now, Siri, Alexa and Cortana, ‘M’ Assistant recognizes text-based natural language only (Wagner). Test users such as Kantrowitz shared their experiences in the media and discussed the strange tasks they assigned to M Assistant – from buying flowers and booking flights to testing M’s creativity to draw and write a poem. Those and other tasks were accomplished thanks to the help of ‘real’ humans who were helping ‘M’ (Statt) and taking over when it wasn’t unable to complete a request, thus training it. More about the testing, training and a discussion on the issue of Big Data you can find in Kersten’s blog post.
Apparently, M is learning slowly but surely and two years after the testing, M was launched publicly in the USA (even though with a much more limited number of features than the beta version – as we can see in the video below).
At the time of writing the features send a sticker, share a location, make a plan, start a poll (in a group chat only), save content, wish Happy Birthday, initiate a voice or video call are available to all users of Messenger in Australia, Canada, South Africa and the United Kingdom (Landowski and El Moujahid). In addition to those, the users in the USA can make payments, order food, book a ride, send GIFs, buy movie tickets and send quick replies. The latest three were just launched on the 21st of September 2017. According to Facebook this is ‘just the beginning’ because M is expected to come to other countries and to keep adding new features.
It is important to note that users can access the features available in their country at any time anyway but when the AI recognizes that they need them, it would join the conversation and suggest using them. Those suggestions are ‘nudges’ which Thaler and Sunstein define as ‘any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options or significantly changing their economic incentives’ (6). Facebook as a choice architect is placing the suggestions in the user interface but the users are free to choose whether to use them or not and what is more, they have the option to mute the assistant. However, according to the scholars people tend to follow the ‘default’ options (Thaler and Sunstein 8), so following their argument, when presented with the nudge to send a Goodnight sticker instead of writing ‘Goodnight’, users might as well do this. Or instead of simply replying ‘I am in the park’, they will share their exact live location. Thus, encouraging users to make use of all available options, Facebook is able to collect even more data and to train ‘M’ even better. How will our communication be affected when we are encouraged to use more and more GIFs, stickers and ‘automated’ quick replies? Is ‘M’ going to foster a new form of communication?
New Territory For Advertising
There are numerous issues that the AI Assistant brings but I would like to analyze ‘M’ using the three ‘narratives of technological utopianism’ which West (17) outlines, connect them to the work of other scholar and thus illustrate how Messenger is becoming a new marketing territory for advertisers. The narratives – value of the free and open network, making the web personal and the technocratic value placed on data (West 17) demonstrate the course of persuasion which is happening in order to enable users to voluntarily share their information and become members of the ‘online communities’. West coins the term – data capitalism which focuses on the distribution of power, given to those who have the ‘access and capability’ (West 1) to interpret and use the collected data.
Value of the free and open network
Messenger is free to use which enabled the app to become popular and essential for communication nowadays. Deciding not to use it limits one’s social interactions and their inclusion in the online community. West analyzes the issue the user is facing when balancing one’s social desire to be part of a community while also protecting their own privacy. For Turow this is a ‘cost-benefit calculation’ (114) – users give up their information because they know they are getting rewards, special offers, discounts, news emailed to their mailbox, thus creating a ‘rewarding relationship’ (119). By using ‘M’ users are getting reminders, suggestions and a quicker and more effective way to book services.
Making the web personal
Messenger’s Product Managers Landowski and El Moudjahid wrote that ‘M learns: the more you use it, the more it can help.’ Who wouldn’t want a personal butler who knows them well and predicts their next move, thus saving them time and effort? By using Messenger every day, we are training ‘M’ which enables it to give us personalized suggestions, based on our language style and usage of specific features. Couldry and Turow discuss microtargeting and native advertising (paid content mimicking the natural one) as ways to target users differently depending on their profiles. Having in mind the fact that Messenger is a member of the big family of Facebook, Instagram, WhatsApp, ‘M’ will easily know enough about us to be able to offer products or services using different personalized persuasion techniques.
By nudging us to use ‘M’, Facebook implements ‘a deeply intentional and highly consequential new logic of accumulation’ – a definition Zuboff uses for her term surveillance capitalism (75). This new logic aims to ‘predict and modify human behavior as a means to produce revenue and market control’ (Zuboff 75). As Zuboff points out Big Data is made up of small data (79), so no matter how insignificant it might sound to you that Facebook knows who do you say goodnight to and know, any data is analyzed and monetized. The more input we give to ‘M’, the better it will be become at predicting our behaviour, thus gaining more and more influence and profit.
Technocratic value placed on data and its potential to augment consumer power
Power is distributed to those willing to buy our data in order to get profit in return (West). Delivery.com, Lyft, Uber and Fandango are excellent examples to look at. If you discuss with a friend that you feel like watching a film, ‘M’ Assistant will pop in and nudge you to book a ticket with Fandango. If it detects that you need to go somewhere, it will nudge you to book a ride. Those companies are buying opportunities to influence us and are thus becoming what Zuboff calls the Big Other. They are implementing a new way of market relationship which ‘M’ enabled.
If (or when) the features available in the beta version (shopping being one) become publicly available, M Assistant will provide a new business model for marketers to advertise. But even if we look at the features available at the moment, ‘M’ already provides a new territory for marketing surveillance, a term used by Turow. And if (or most probably once) ‘M’ starts to nudge us to do our shopping and book the events we want to attend through Messenger, all we will have to do is choose between the few options we are given and of course – pay for it. But Facebook has already made some choices for us- if we decide to choose the default, we can go to the Fangando cinema with Uber, after having our lunch Delivery. The issue Couldry and Turow (1712) highlight is that in the advertising industry choices are made for us using criteria we are not aware of.
‘M’ assistant will change the way marketers advertise because every new territory brings new rules. Our private conversations are listened to and companies are helpfully waiting to suggest the ‘best’ products and services to us. ‘M’ has already entered our living room, so are we then going to need the Web once we have everything there?
Kantrowitz, Alex. Facebook Reveals The Secrets Behind “M,” Its Artificial Intelligence Bot. 19 November 2015. 24 September 2017. <https://www.buzzfeed.com/alexkantrowitz/time-to-meet-the-wizard-facebooks-messenger-head-pulls-back?utm_term=.no70KaRry#.ruBazkG3K>.
Kersten, Lianne. Facebook’s new virtual assistant ‘M’: mastermind or menace? 15 Sep 2015. 24 Sep 2017. <https://mastersofmedia.hum.uva.nl/blog/2015/09/15/facebooks-new-virtual-assistant-m-mastermind-or-menace/>.
Laurent Landowski, Kemal El Moujahid. M Now Offers Suggestions to Make Your Messenger Experience More Useful, Seamless and Delightful. 6 April 2017. 24 Sep 2017. <https://newsroom.fb.com/news/2017/04/m-now-offers-suggestions-to-make-your-messenger-experience-more-useful-seamless-and-delightful/>.
Marcus, David. Messenger Brings People Together All Over the World. 18 April 2017. 24 September 2017. <https://newsroom.fb.com/news/2017/04/messenger-f8/>.
Nick Couldry, Joseph Turow. “Advertising, big data and the clearance of the public.” International Journal of Communication 8 (2014): 1710-1726.
Richard H. Thaler, Cass R. Sunstein. Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven & London: Yale University Press, 2008.
Statt, Nick. Hands-on with Facebook M: the virtual assistant with a (real) human touch. 26 October 2015. 24 September 2017. <https://www.theverge.com/2015/10/26/9605526/facebook-m-hands-on-personal-assistant-ai>.
Turow, Joseph. “Audience Construction and Culture Production: Marketing Surveillance in the Digital Age.” The ANNALS of the American Academy of Political and Social Science (2005): 103-121.
Wagner, Kurt. Here’s why Facebook Messenger isn’t building a voice-controlled assistant like Alexa or Siri. 2 May 2017. 24 Sep 2017. <https://www.recode.net/2017/5/2/15525048/facebook-messenger-m-voice-control-assistant>.
West, Sarah Myers. “Data Capitalism: Redefining the Logics of Surveillance and Privacy.” Business & Society (2017): 1-22.
Zuboff, Shoshana. “Big other: surveillance capitalism and the prospects of an information civilization.” Journal of Information Technology (2015): 75–89.