Dystopian Narratives to �(re)think AI

Ilaria Tiddi

Vrije Universiteit Amsterdam

03/11/2022

Warm up

Black Mirror

• Sci-fi British TV series by Charlie Brooker
• 5 seasons, 2011 – today
• Speculative fiction : near-future, dystopian scenarios
• Focus
• Dark and satirical themes examining modern society
• Unanticipated consequences of new technologies

Black Mirror Episodes

Which episodes do you remember (scary, catchy…)?

Black Mirror Episodes

Which episodes do you remember (scary, catchy…)?

• 15M merits : riding in exchange of merits
• Be right back : an AI imitation of a dead boyfriend
• Nosedive : eye implants and mobile devices for social interactions rating
• Hated in the Nation : hate-speech on social media becomes deadly
• Metalhead : a warehouse dog chasing thieves �of children goodies

Why do we remember these episodes?

• Existing technologies
• Fictional situations
• Speculation on how far before them becoming real

Boston Dynamics robotic dogs

US/China benefit systems

LuKa : Chatbots for the deads

Why do we remember these episodes?

In other words, dystopian narratives

Boston Dynamics robotic dogs

US/China benefit systems

LuKa : Chatbots for the deads

Black Mirror : main technique

Visual Vignettes (i.e. short narratives)

• A social science method
• Using drawn vignettes for qualitative research assessment
• combines the vignette and the photo essay genres
• multimodal format that encourages analysis and critical thinking
• opportunity to critically engage in and rethink a topic using visuals and text

A research hypothesis

Given

• Misuse of technology
• Current debate on data ethics �and algorithmic governance
• The VV method �and BM experience

Can we use dystopian narratives (and science fiction) to help critical thinking about AI ethical issues?

https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing

Recoding Black-Mirror (2017–2019)

Audience

• Computer/data scientists, social scientists, privacy advocates
• interest in AI ethical aspects
• AI role in constructing our future

Topics

• Ethical/societal challenges of digital technologies
• Computer Science solutions against the misuse of technologies
• Technology to prevent Black Mirror’s future.

https://kmitd.github.io/recoding-black-mirror/rbm-2017.html�https://kmitd.github.io/recoding-black-mirror/rbm-2018.html�https://kmitd.github.io/recoding-black-mirror/cpdp2019.html

Recoding Black-Mirror (2017–2019)

Goal

• Push to consider potential ethical / social risks of technologies
• Offer possible solutions

Method

• Written dystopian scenarios (extends VV)
• based on one own’s research
• Inspired by Black Mirror episodes

https://kmitd.github.io/recoding-black-mirror/rbm-2017.html�https://kmitd.github.io/recoding-black-mirror/rbm-2018.html�https://kmitd.github.io/recoding-black-mirror/cpdp2019.html

Some Recoding BlackMirror Examples

• WannaDie
• a new ransomware threatening humans for counterpayments
• “pay or your pacemaker will get desactivated”
• cyber-health threats, IoT-connected word
• Brain-Computer Interfaces
• neurosecurity issues and privacy implications of BCIs
• Mindreading in dystopic scenarios
• Academia 4.0
• academic CVs in exchange of merits

Some Recoding BlackMirror Examples

• Internet of Neurons
• a router captures our brainwaves
• humans as a ‘thing’ in the Internet of Neurons
• Surveillance
• email surveillance of emotions at work
• Which moral rights does privacy at work have?
• Shut up and run
• the never-ending quest for social fitness
• negative drawbacks in the use of wearable sensors

Recoding BlackMirror Hands-on

• interactive exercise
• split into participants in groups
• script a new Black Mirror-inspired episode
• futuristic transport (bikes, cars)
• dystopian consequences of transport control
• Ethical divide between those who embrace technology and those that do not

Thank you

Questions?�i.tiddi@vu.nl�@IlaTiddi