From Typewriter to AI

The changing face of media and its ethical challenges.

In the 21st century its one thing to come up with a story but its another thing to come up with a way of telling that story. Storytelling is the most essential kind of human trait and in my opinion can only be told from a human perspective. But there are now many different platforms that allow for stories to be told freely by anyone, or it seems, by anything.

I could wax lyrical about the history of storytelling from stone age cave paintings onwards, but Id probably bore myself writing it as much as subjecting you to read it. There will be one of those funky infographic timeline pictorials featured later on in this conversation that’ll probably summarise that, as soon as I find the time to make one. The way we tell stories has changed mostly due to the emergence of technology that enables the democratisation of communicating to the world. Podcasts in particular give virtually anyone the ability to tell stories in the most human simple and fundamental ways. Telling stories in the spoken word is very personal, and somewhat comforting, as it probably was pre 1940s when listening to a radio. Devised by apple in 2005 podcasting has seen a meteoric rise in popularity particularly in the US where it reaches millions of listeners. According to an Edison Research survey, US audiences listened to an average of 5 podcasts a week totalling 4 hours on average and 24hours after publication. By 2016 there were 98 million Americans listening ( Perez-Latre, 2018).

Podcasting surged to celebrity status around 2015/16 with luminaries such as Arnold Schwarzenegger popping in on blogger Tim Ferriss’s show having a very candid interview. Ferriss has a quoted 350 million downloads of his show, with The Observer and other media have naming him “the Oprah of audio” due to the influence of his podcasts. And also you’ve got to love Arnie.

Barack Obama being interviewed with podcaster Marc Maron (June 2015) to discuss everyday things like college, fitting in, race relations, gun violence, changing the status quo, disappointing your fans, comedians, fatherhood and overcoming fear.

The same can be said for Vlogging and the surge of YouTubers armed with either theyre influencing sense of something everyone wants, or to offer advice and opinions on an interesting subject. All basically wrapped up in the art of communicating this from a personal level through storytelling. There is also something to be said about the democratisation of broadcasting through Podcasting and Vlogging.

It is however becoming more and more apparent, as we engage with new technology that a good deal of data about us is being captured, either through social media, IoT devices or internet applications. There are many challenges to this, the most immediate one being what is done with this information and how can we make sense of it. How do we find the time to sift through it effectively to get answers, results, ideas, or convey a message? This has led to the evolution of software applications that learn our daily habits through the process of ML (Machine Learning). Newsrooms across Europe and the US are coming up with many automated ways of recovering large amounts of statistical data on sports results, elections, company earnings, stock etc. As pointed out by the director of News Partnerships at the US led Associated Press Lisa Gibbs

“Earnings and sports were obvious for us because they’re data driven, The sources of data are clean, and there is a value of having information about them out very quickly.” Delcker, (March 2019)

Reuters news agency is also picking up on this in a big way, where they’re building a “cybernetic newsroom” combining the machine capability of trawling through and collating masses of data, with human judgement. An AI tool which they’re developing called Lynx Insight identifies trends, facts and suggests new stories to write. At the same time journalistic algorithms written into the software encourage a data driven approach to the story. Lynx effectively “ leverages the smarts of our newsroom, both in asking the right questions of the machines and in evaluating the answers that come back, to drive even better journalism, and much more quickly.” Chua (March 2018).

So how would this automated Journalism work? Say for example, the Pennine Moors brush fires that happened last year could have been covered by fire-proofed drones with cameras covering minute by minute aspects of the fire, plotting its direction and seeing close hand how its being dealt with, non-stop for the full duration of the incident. All this data would be collated by an AI that would predict which conurbation would be hit next, and with relevant footage. News chatbots would be available online to answer public questions of the event and how it effects them.

However whilst the blue sky thinkers amongst us see AI as the answer to all our problems and a means of enhancing our human accident and failure-prone characteristics, I remain cautious and rather concerned about this new technical revolution. It is true that learning algorithms and automated neural networks greatly enhance our means to acquire accurate information, but these aren’t the amazing thinking and learning machines that the general media make them out to be. As Tomas Mikolov a research scientist at Facebook AI points out, our best machine learning algorithms are merely tools which collate masses of past information and run statistical models against it. This is far from “learning” as humans do. AI can be trained to do something, but that doesn’t mean it will understand why its doing it. Robitzski (October,2018). So really we aren’t going to see Skynet taking over the world any time soon. Or are we?

Whatsmore the current version of AI cannot tell a good story, a story that moves or inspires us as AI has no empathy in a way that touches us emotionally. The main area of my concern with News/Media Storytelling AI is our control of our own personal data, or lack of. It has in fact taken a number of international scandals including Facebook\Cambridge Analytica to highlight just how much we don’t know about where our data goes and whats being done with it. And there’s the arrogant aspect of it all which is indemic in our society of the haves and have nots in knowledge. Just because a company has some spurious personal data about me, that doesn’t give them permission or authority to comment on the way I live my life and then expect me to thank them for it and stand in awe of their greatness. There is also a danger that data is captured by only a certain demographic of educated, Industrialised, well off and mostly white people, giving a skewed perception of what any AI could deliver. Its for these and other reasons that 2018 was a challenging year for the AI.

The positive out of this is that designers and implementers are starting to consider the ethics of dealing with peoples personal information aspects of their lives. Phil Hesketh, a designer and researcher has developed an ethics toolkit that is an online “ethics checker” that asks questions throughout every step of an project on whether you are considering the position empathy or cultural leanings of the particular public demographic your communicating with. One of many ethical checkers that are in development.

So to conclude, as media progresses to automation and access to more highly intuitive content that is catered to your personal needs and perspectives, new challenges have arisen that should be addressed. These exciting times must match drive towards personalised media information with a drive towards ethical practice and regulation.

References

The Interactive Advertising Bureau (IAB) IAB-Edison Research Podcast Advertising Study. [online] IAB. Available at: https://www.iab.com/wp-content/uploads/2016/09/IAB-Edison-Research-Podcast-Advertising-Study_Updated.pdf/ [Accessed 2nd May, 2019].

Perez-Latre, F.J. (Sept, 2016) The Rise of Podcasting and Trust in Legacy Media: Opportunities and Challenges for the Mobile “Radio of the Future”

Delcker, J. (2019). This story was not written by a robot. [online] POLITICO. Available at: https://www.politico.eu/article/robot-reporters-newsroom-algorithms-artificial-intelligence/ [Accessed 2nd May, 2019].

Ferris, T. (Feb, 2015) Tim Ferriss Interviews Arnold Schwarzenegger on Psychological Warfare (And Much More) (#60). [online]Available at: https://tim.blog/2015/02/02/arnold-schwarzenegger/ [Accessed 2nd May, 2019]

Chua, R. (March 2018) The cybernetic newsroom: horses and cars.[online] REUTERS.Available at: https://www.reuters.com/article/rpb-cyber/the-cybernetic-newsroom-horses-and-cars-idUSKCN1GO0Z0 [Accessed 4th May, 2019]

Robitzski,D. (October 2018) You Have No Idea What Artificial Intelligence Really Does. [online] FUTURISM. Available at: https://futurism.com/artificial-intelligence-hype [Accessed 5th May 2019]


Originally published at https://medium.com on May 7, 2019.