ARIN6905 – Digital Creators and Audiences - Week 5
Algorithms, AI, and
Audiences
Algorithms
A set of rules followed by a computing
system to create some sort of outcome
Ubiquitous in computer software,
comparable in a sense to mathematical
proofs – where data can be inserted and an
outcome will be generated in a consistent
manner according to the data added
However, as with all digital technology, only
as “effective” or “good” as its design
AI
A heavily overused phrase which describes a
genuine field of research and implementation, but
also has become what we call an “empty signifier”,
a phrase which can simply be deployed to a certain
effect without truly meaning anything
Generally used to describe some kind of non-human or non-biological agent (this is
inevitably vague) that displays behaviour that might be seen to be “intelligent”
(again, this is vague and subjective)
AI can be seen as a more involved version of algorithmic thinking, or a vast
combination of algorithms, or something able to produce results not necessarily
immediately anticipated
Algorithms, AI, and Audiences
Algorithms and AI systems are increasingly reshaping how we access online
content and what sort of content is presented to us
Think about how websites like Netflix or Facebook prioritise what you see
These are not decisions made by humans, nor by random, but rather by
systems attempting to figure out what content would be meaningful to you
Central point: this is not a neutral process, but one which is complicated by
the algorthims / AI themselves, technologies, design intentions, platform
specifics, metrics and definitions, and user experience
Hallinan, B., & Striphas, T. (2016).
Recommended for you: The Netflix Prize and the
production of algorithmic culture
The “Netflix Prize”
An example of a company looking to “crowdsource”
improvements to its algorithms
This is surprisingly common – many tech companies, for example, offer “bounties” for
people who can report glitches or bugs or vulnerabilities in their systems
The Netflix Prize offered $1m to a team who could demonstrate a significant
improvement over what Netflix’s system was already doing
Take note, of course, that Netflix decided how to “rate” systems and Netflix set the
goal posts, by defining what they were looking for
The “Netflix Prize”
Offered a rare chance to “scrutinise” the
algorithms which underpin Netflix
As Hallinan & Striphas note, these are generally kept extremely secret, closely protected,
and are often central parts of a company’s “IP” (intellectual property)
There are many obvious (financial and strategic) reasons for this…
…and yet it also prevents oversight, understanding, and any kind of critical analysis
(Algorithmic) culture
“What does culture mean, and what might it be coming to mean, given the growing
presence of algorithmic recommendation systems”?
“The use of computational processes to sort, classify, and hierarchize people, places,
objects, and ideas, and also the habits of thought, conduct, and expression that arise in
relationship to those processes”
“A gradual shift away from debates about great works, or defining canons, to something
like the opposite: how to moderate elements of the cultural field that may present
themselves as atypical or outstanding, so that they can be led to make sense relative to
other, more even-keeled, examples”
(Algorithmic) culture
Such algorithms reshape the culture we consume; how
we think about that culture; what culture we even see;
and what culture is prioritized
“We argue that although the Netflix Prize may have been billed as a competition to improve
the company’s recommendation algorithm, it was equally an effort to reinterpret what
culture is—how it is evaluated, by whom, and to what ends.”
Ambiguity between what people think a recommendation system is doing and what it is really
doing. How are ratings actually kept? Do they all have equal value? Are hidden things also
being factored in?
Exercise#1
How accurately do you think
recommendations systems on major
platforms – Netflix, YouTube, Twitter,
Facebook, etc – represent your interests?
Do you ever see things that seem totally
unrelated to you? Why do you think these
sometimes appear?
Burroughs, B. (2019). House of Netflix: Streaming
media and digital lore. Popular Communication, 17(1),
1-17.
Streaming video as disruption
How is are the television and film industries approaching the rise of Netflix, Amazon
Video, and other online video streaming services?
As the paper notes, controversies or major changes can often render “visible” social,
political or especially economic relationships that are otherwise normally hard to see
“Streaming companies have come to embody the strategic power of Internet companies
(the audience tactic re-articulated as a strategy), which are butting heads with other
regimes of corporate power”
“Cable and telecommunication companies such as Comcast and Time
Warner Cable are dealing with the loss of subscribers and anxieties about
“cord-cutters” and “cord-nevers” amplified by the growth, popularity, and
cultural salience of streaming services such as Netflix”
Industry lore as “the conventional knowledge among industry insiders
about what kinds of media culture are and are not possible, and what
audiences that culture will and will not attract” (Havens, 2008)
There is no established body of knowledge – although this is growing
constantly – about video streaming platform audiences
Streaming video as disruption
Streaming video as disruption
Video streaming services are hence disrupting this industry…
…and algorithms and AI are playing a central role. They enable these platforms to (with at
least some success) focus on the content viewers want to see, which – when coupled with
the immediacy of accessible content – is a major challenge to traditional TV
What previously occupied some of these algorithmic roles? Critics, magazines, schedules,
friends, etc
Exercise #2
Have you become a cord-cutter? Are you
even a cord-never?
Do you think you might move in that
direction in the future? If so / if not, why?
Is anyone in your family more or less
likely to cord-cut? Why?
Siles, I., Espinoza-Rojas, J., Naranjo, A., & Tristán, M.
F. (2019). The mutual domestication of users and
algorithmic recommendations on Netflix.
Communication, Culture & Critique, 12(4), 499-518.
Domestication of users and platform
Five ways a mutual domestication takes
place:
• “personalization, or the ways in which
individualized relationships between
users and the platform are built; “
• “how algorithmic recommendations
are integrated into a matrix of cultural
codes;”
Domestication of users and platform
• “the rituals through which they are
incorporated into spatial and temporal
processes in daily life; “
• “the resistance to various aspects of Netflix
as a form to enact agency; “
• “and the conversion or transformation of the
private consumption of the platform into a
public issue “
(Exercise#3)
What do you think about the idea of TV
shows “designed” solely by humans, vs
those co-designed by humans and
algorithms? Do you have a preference for
one or the other? Would you watch a TV
show or film heavily “written” by an AI?
Algorithms, AI, and Audiences
The audience experience of online content is being transformed by the roles
played by algorithmic and AI systems
These shape what we see, how and when we see it, how we consume it, what
options are presented to us for consumption, and how consuming online content
fits into the rest of our lives and our routines and rhythms
Netflix is only one example – for your case studies or presentations you can
easily consider almost any other major online platform
Thanksfor
theclass!

学霸联盟