The Role of Bots in Live Streaming [Miller 2019]

Gagen and Cook’s description of the concept of liveness on Second Life gave insight into the difficulties in
defining a virtual act as ‘live’ due to the constraints of virtual space. They touched upon a particular view from
user Anya Ristow who described the problem of bots in Second Life, and felt that in a ‘live’ concert the
presence of bots meant “the concert was not the live, communal event it made itself out to be but rather a
fraud” (Gagen and Cook 198). This concept led their argument that to present music live on Second Life “is to
enter a kind of social contract” which is determined not only by the performer but by the listener/audience
(Gagen and Cook 199). In current day, acts of live streaming such as Bryce Xavier on live.ly, bots continue to
be present in audiences. The question I’m interested in is: does the presence of bots in the audience impact
how we define live streaming?

Within this framework of the social contract we can separately define the roles of the live streamer/influencer
and the audience. The influencer’s role is to produce content, and the audience’s role is to listen and react to
the content. Using the earlier framework we studied in class from Crawford, the audience is engaging in
reciprocal listening. It is the presence of the audience’s reciprocal listening which gives a live streamer their
influence. Live streaming tools such as live.ly, instagram live, and twitch show how many listeners are tuned
in and users can make comments which will appear publicly in real time. This information directly impacts the
content produced by the live streamer. In the article about Bryce Xavier, we see various ways in which users
can react and even pay money to an influencer for the chance to for their comments to be acknowledged and
reciprocated (Hess). 

I think a really interesting way of looking at bots in this respect is again through Crawford’s framework, but
this time as delegated listening. In class we talked a lot about delegated listening in the context of corporations
delegating interns to respond to twitter messages, AKA delegating for the producer. I instead want to think of
how bots exist as delegated listening for the audience. If the role of the audience is to react to the content, then
can we think of bots as delegated listening from the audience? This would create an important difference in
how ‘live’ a live stream is. If the audience’s role is being delegated to a non-real person, then the interaction
may no longer fit inside of ‘real time’ because there is a non-real interaction occurring. While bots exist on
many social media platforms, they more heavily impact the act of live streaming because of the real time
implications. Reactions from bots can misshape how the live streamer perceives the audience and can therefore
affect the content. With no real time interaction, there can be no ‘live’ interaction, thus defeating the purpose
of the live stream.
Image result for instagram bots

Works Cited

Crawford, Kate. "Following You: Disciplines of Listening in Social Media." The Sound Studies
Reader. By Jonathan Sterne. London: Routledge, 2012. 79-90. Print.
Gagen, Justin, and Nicholas Cook. "Performing Live in Second Life." OXFORD HANDBOOK OF
MUSIC AND VIRTUALITY. By SHEILA WHITELEY. S.l.: OXFORD UNIV US, 2019. 191-209.
Print.
Hess, Amanda. "The Teenage Life, Streamed Live and for Profit." The New York Times. The New York
Times, 06 June 2017. Web. 05 Nov. 2019.







Comments

Maggie said…
I think it's also interesting to think about how the "labor" of these bots (or maybe lack thereof? Since these non-human bots are doing work audience members would otherwise do, does it still count as labor?) may be financially beneficial for these livestreamers. Often times, online influencers' popularity is garnered from viewers' interactions, and that is what dictates the economic profit they gain from their online action. So it would seem reasonable to gather that one role of bots in livestreaming is to create profit for livestreamers. However, since these bots are not real people and therefore cannot directly depict how real individuals would react to these livestreams, would it be right (or ethical) to reward the livestreamers they help out? This also makes me question whether or not livestreamers can pay to have bots react to their videos (I would assume yes), and how this challenges the ethics of our consumer economy.
Elaine Kim said…
The question that I have when I think of ‘live’ performances in Second Life and delegated listening through bots is: Does delegated listening require an individual to delegate the task of listening to a human being? Does the ‘thing’ we delegate the task of listening to have to be autonomous, sentient, and alive? Or, as you suggest, can we delegate listening to an inanimate object or, in this case, avatar? I would argue that the ‘thing’ that we delegate listening to has to be able to perform the task of listening (hearing and interpreting is heard) autonomously. Therefore, I do not think I would consider the act of listening to ‘live’ performances on Second Life delegated listening. Second life avatars rely upon users for its existence; the avatar would not exist without a user to control it and move it around in the virtual world. The avatar is an extension of the self, not a separate being that individuals delegate tasks to.
Carrie Phillips said…
Something about a delegated bot audience is even more eerie than a "delegated producer." As a performer myself, I respond and adapt to the audience. I am under the impression that reactions and behaviors before me are authentic. Granted this discussion of bots applies more to virtual spaces as opposed to physical spaces. However, I don't believe this authenticity has to be isolated to the physical world, and believe that manipulation of both audience and producer by delegated bot listening is particularly dangerous. Content production could become completely a result of bot algorithms as opposed to visceral, emotional reactions by audience members.