SlideShare a Scribd company logo
1 of 27
Download to read offline
Data Ethics
Mathieu d’Aquin - @mdaquin
Data Science Institute
National University of Ireland Galway
Insight Centre for Data Analytics
Data Ethics
Data Ethics
The set of principles and processes that guide the ethical
collection, processing, analysis, use and application of data having
an effect on human lives and society
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Data Ethics
The set of principles and processes that guide the ethical
collection, processing, analysis, use and application of data having
an effect on human lives and society
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Data Ethics
The set of principles and processes that guide the ethical
collection, processing, analysis, use and application of data having
an effect on human lives and society
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Data Ethics
The set of principles and processes that guide the ethical
collection, processing, analysis, use and application of data having
an effect on human lives and society
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Ethics
What is right, what is fair, what is just.
Hosmer, L. T. (1995). "Trust: The Connecting Link between Organizational Theory and Philosophical Ethics". The
Academy of Management Review. 20 (2)
In an ideal world
What is ethical.
(right, fair, just)
What is legal.
In the real world
What is ethical.
(right, fair, just)
What is legal.
What does this
have to do with
data?
What is ethical.
(right, fair, just)
What is legal.
What does this
have to do with
data?
Data protection
Privacy Statistical bias
Black box decisions
Uneven access
self-governance
...
Machine ethics
(https://www.smbc-comics.com/comic/machine-ethics)
Example related to privacy/data protection
In 2014, New York City released data about 173m taxi
trips in the city, where the licence plates and identifier of
the taxi had been obfuscated for anonymisation
purposes.
It was de-anonymised within hours of being released…
… and later cross-referenced with timestamped pictures
of celebrities entering taxis in New York to figure out their
personal address, and how much they tipped.
See e.g. http://gawker.com/the-public-nyc-taxicab-database-that-accidentally-track-1646724546
Example related to privacy/data protection
In this case, it is useful to note that:
- Replacing identifiers with a hash is not anonymisation, it is
at best bad pseudonymisation
- Current data protection regulation in Europe regulates
against this sort of cases
- The upcoming GDPR will make the consequences of this
sort of mistakes stronger
- It defines its scope as “any information relating to an
identified or identifiable natural person ('data subject'); an
identifiable natural person is one who can be identified,
directly or indirectly”. Arguably, the unanticipated case of
the celebrities fall under this scope… and should therefore
have been anticipated.
But, should also
be asking:
What is my impact
on society? How can
I minimise the risk
of negative
implications?
(drawing upon critical
social science, and
regulation as guidelines)
How do I make
what I’m doing
compliant with
regulation?
In addition to:
Examples related to bias
Google search “unprofessional hair for work” and
“professional hair for work”
Example related to black-box decision
The US justice system relies on a tool to predict, when
judging for an offence, what is the likeliness an individual has
to re-offend.
It is based on many variables, including address, type of
offence, past history of offences, and ethnicity.
It has been demonstrated to make significant mistakes,
especially through being prone to give overly negative scores
to black people.
See https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing
Notes on those cases
- The algorithm is not biased, the data is. Garbage in,
garbage out.
- Human decisions are not gold standards, and therefore
should not be treated as such in training machine
learning models
- Sometimes, unrelated things just happen to correlate
(see http://www.tylervigen.com/spurious-correlations) - a
machine learning model will rely on those correlations to
make decisions.
Should we ban cheese?
Example related to uneven access and
under-represented cases
Researchers at Georgia Institute of Technology
developed and used a chatbot to act as a TA for
computer science courses (without the students’
knowledge).
It worked very well in most cases…
… but failed dramatically in uncommon, delicate
situation.
Bobbie Eicher et al., Jill Watson Doesn’t Care if You’re Pregnant:
Grounding AI Ethics in Empirical Studies, AIES 2018
Example related to uneven access and
under-represented cases
Notes on this case:
- Another form of bias, not related to spurious
or inaccurate correlations, but to
under-representation of specific parts of the
population.
- Raise issues with the uneven access to the
benefit of the technology, and therefore
unfairness.
- “The future is already here — it's just not very
evenly distributed” -- William Gibson
Bobbie Eicher et al., Jill Watson Doesn’t Care if You’re Pregnant:
Grounding AI Ethics in Empirical Studies, AIES 2018
Principles for designing ethics data science projects
‘Ethics in
Design’ for Data
Science
Dialectic
The process is based on a conversational
approach between data and critical social
scientists throughout the project’s life-cycle.
Reflective
Ethical concerns are not pre-fixed; they may
emanate from any stage of the project; thus,
constant reflexivity on activities and
researchers is needed.
Creative, not disruptive
The objective of this process is to achieve a
positive impact on the research, increase its
value addressing ethics throughout the
project’s life-cycle.
All- encompassing
Ethical concerns appear as much in the
research activities as in their outcomes, their
use and exploitation; the process needs to
expand on all stages.
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Principles for designing ethics data science projects
‘Ethics in
Design’ for Data
Science
Dialectic
The process is based on a conversational
approach between data and critical social
scientists throughout the project’s life-cycle.
Reflective
Ethical concerns are not pre-fixed; they may
emanate from any stage of the project; thus,
constant reflexivity on activities and
researchers is needed.
Creative, not disruptive
The objective of this process is to achieve a
positive impact on the research, increase its
value addressing ethics throughout the
project’s life-cycle.
All- encompassing
Ethical concerns appear as much in the
research activities as in their outcomes, their
use and exploitation; the process needs to
expand on all stages.
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Methodology borrowed from design fiction:
the use of speculative and often provocative
scenarios involving the artifact to be design (a
data process), as a way to explore its
possible implications and reflect on their
consequences.
Pragmatically, it consist in telling stories
asking and answering what if questions (e.g.
“what if the student is pregnant? What would
happen then?”) and building mockups of the
final product to reflect on its behaviour.
See Anthony Dunne and Fiona
Raby, Speculative Everything, MIT
Press, 2013
and
Joseph Lindley and Paul Coulton,
"Back to the Future: 10 Years of
Design Fiction". British HCI 2015.
Principles for designing ethics data science projects
‘Ethics in
Design’ for Data
Science
Dialectic
The process is based on a conversational
approach between data and critical social
scientists throughout the project’s life-cycle.
Reflective
Ethical concerns are not pre-fixed; they may
emanate from any stage of the project; thus,
constant reflexivity on activities and
researchers is needed.
Creative, not disruptive
The objective of this process is to achieve a
positive impact on the research, increase its
value addressing ethics throughout the
project’s life-cycle.
All- encompassing
Ethical concerns appear as much in the
research activities as in their outcomes, their
use and exploitation; the process needs to
expand on all stages.
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
Principles for designing ethics data science projects
‘Ethics in
Design’ for Data
Science
Dialectic
The process is based on a conversational
approach between data and critical social
scientists throughout the project’s life-cycle.
Reflective
Ethical concerns are not pre-fixed; they may
emanate from any stage of the project; thus,
constant reflexivity on activities and
researchers is needed.
Creative, not disruptive
The objective of this process is to achieve a
positive impact on the research, increase its
value addressing ethics throughout the
project’s life-cycle.
All- encompassing
Ethical concerns appear as much in the
research activities as in their outcomes, their
use and exploitation; the process needs to
expand on all stages.
d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
i.e. don’t do that:
Some conclusions
Following regulation is insufficient for data ethics.
Ethical issues often appear after the development
phase, in scenarios that have not been
anticipated.
Need to uncover those scenarios to integrate in
the process ways of mitigating ethical
implications, and balance social, economic and
ethical values.
This cannot be done (currently) by the
technologists alone!
Shameless self-promotion
Check
Towards an “Ethics by Design” methodology for AI research projects at the first
conference on AI, Ethics and Society, AIES 2018
The Re-Coding Black Mirror worksop at The Web Conference (WWW 2018) -
https://kmitd.github.io/recoding-black-mirror/
MagnaCartaForData.org
Contacts: mathieu.daquin@insight-centre.ie, mdaquin.net, @mdaquin

More Related Content

What's hot

Data preprocessing
Data preprocessingData preprocessing
Data preprocessing
ankur bhalla
 

What's hot (20)

Data preprocessing
Data preprocessingData preprocessing
Data preprocessing
 
Data quality and data profiling
Data quality and data profilingData quality and data profiling
Data quality and data profiling
 
Web mining
Web miningWeb mining
Web mining
 
Data Cleaning Techniques
Data Cleaning TechniquesData Cleaning Techniques
Data Cleaning Techniques
 
Data Science
Data ScienceData Science
Data Science
 
Data science applications and usecases
Data science applications and usecasesData science applications and usecases
Data science applications and usecases
 
Big Data: Issues and Challenges
Big Data: Issues and ChallengesBig Data: Issues and Challenges
Big Data: Issues and Challenges
 
CRISP-DM: a data science project methodology
CRISP-DM: a data science project methodologyCRISP-DM: a data science project methodology
CRISP-DM: a data science project methodology
 
Big Data Ecosystem
Big Data EcosystemBig Data Ecosystem
Big Data Ecosystem
 
Big data Presentation
Big data PresentationBig data Presentation
Big data Presentation
 
Data Wrangling
Data WranglingData Wrangling
Data Wrangling
 
Big data case study collection
Big data   case study collectionBig data   case study collection
Big data case study collection
 
Web mining
Web mining Web mining
Web mining
 
Introduction to Data Analytics
Introduction to Data AnalyticsIntroduction to Data Analytics
Introduction to Data Analytics
 
Data Analytics
Data AnalyticsData Analytics
Data Analytics
 
Data Analytics For Beginners | Introduction To Data Analytics | Data Analytic...
Data Analytics For Beginners | Introduction To Data Analytics | Data Analytic...Data Analytics For Beginners | Introduction To Data Analytics | Data Analytic...
Data Analytics For Beginners | Introduction To Data Analytics | Data Analytic...
 
Big data
Big dataBig data
Big data
 
Big data visualization
Big data visualizationBig data visualization
Big data visualization
 
introduction to data mining tutorial
introduction to data mining tutorial introduction to data mining tutorial
introduction to data mining tutorial
 
Data Mining
Data MiningData Mining
Data Mining
 

Similar to Data ethics

Ai Now institute 2017 report
 Ai Now institute 2017 report Ai Now institute 2017 report
Ai Now institute 2017 report
Willy Marroquin (WillyDevNET)
 
Ethics and Responsible AI Deployment.pptx
Ethics and Responsible AI Deployment.pptxEthics and Responsible AI Deployment.pptx
Ethics and Responsible AI Deployment.pptx
Petar Radanliev
 
Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...
Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...
Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...
Glenn Villanueva
 

Similar to Data ethics (20)

Generative AI - Responsible Path Forward.pdf
Generative AI - Responsible Path Forward.pdfGenerative AI - Responsible Path Forward.pdf
Generative AI - Responsible Path Forward.pdf
 
Ai Now institute 2017 report
 Ai Now institute 2017 report Ai Now institute 2017 report
Ai Now institute 2017 report
 
The Ethical Maze to Be Navigated: AI Bias in Data Science and Technology.
The Ethical Maze to Be Navigated: AI Bias in Data Science and Technology.The Ethical Maze to Be Navigated: AI Bias in Data Science and Technology.
The Ethical Maze to Be Navigated: AI Bias in Data Science and Technology.
 
IA in the Age of AI: Embracing Abstraction and Change at IA Summit 2018
IA in the Age of AI: Embracing Abstraction and Change at IA Summit 2018IA in the Age of AI: Embracing Abstraction and Change at IA Summit 2018
IA in the Age of AI: Embracing Abstraction and Change at IA Summit 2018
 
Bias in algorithmic decision-making: Standards, Algorithmic Literacy and Gove...
Bias in algorithmic decision-making: Standards, Algorithmic Literacy and Gove...Bias in algorithmic decision-making: Standards, Algorithmic Literacy and Gove...
Bias in algorithmic decision-making: Standards, Algorithmic Literacy and Gove...
 
DATAIA & TransAlgo
DATAIA & TransAlgoDATAIA & TransAlgo
DATAIA & TransAlgo
 
Ethics and Responsible AI Deployment.pptx
Ethics and Responsible AI Deployment.pptxEthics and Responsible AI Deployment.pptx
Ethics and Responsible AI Deployment.pptx
 
Research methods - ethics
Research methods - ethicsResearch methods - ethics
Research methods - ethics
 
Data science and ethics in fundraising
Data science and ethics in fundraisingData science and ethics in fundraising
Data science and ethics in fundraising
 
Data Science - An emerging Stream of Science with its Spreading Reach & Impact
Data Science - An emerging Stream of Science with its Spreading Reach & ImpactData Science - An emerging Stream of Science with its Spreading Reach & Impact
Data Science - An emerging Stream of Science with its Spreading Reach & Impact
 
Creating a new culture around authenticity and generative AI
Creating a new culture around authenticity and generative AICreating a new culture around authenticity and generative AI
Creating a new culture around authenticity and generative AI
 
[DSC Europe 23] Bunmi Akinremi - Ethical Considerations in Predictive Analytics
[DSC Europe 23] Bunmi Akinremi - Ethical Considerations in Predictive Analytics[DSC Europe 23] Bunmi Akinremi - Ethical Considerations in Predictive Analytics
[DSC Europe 23] Bunmi Akinremi - Ethical Considerations in Predictive Analytics
 
Philosophical Aspects of Big Data
Philosophical Aspects of Big DataPhilosophical Aspects of Big Data
Philosophical Aspects of Big Data
 
Fundamentals of Data science Introduction Unit 1
Fundamentals of Data science Introduction Unit 1Fundamentals of Data science Introduction Unit 1
Fundamentals of Data science Introduction Unit 1
 
e-SIDES workshop at BDV Meet-Up, Sofia 14/05/2018
e-SIDES workshop at BDV Meet-Up, Sofia 14/05/2018e-SIDES workshop at BDV Meet-Up, Sofia 14/05/2018
e-SIDES workshop at BDV Meet-Up, Sofia 14/05/2018
 
AI Governance and Ethics - Industry Standards
AI Governance and Ethics - Industry StandardsAI Governance and Ethics - Industry Standards
AI Governance and Ethics - Industry Standards
 
Data Science-1 (1).ppt
Data Science-1 (1).pptData Science-1 (1).ppt
Data Science-1 (1).ppt
 
The Ethics of AI
The Ethics of AIThe Ethics of AI
The Ethics of AI
 
Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...
Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...
Research Ethics and Integrity | Ethical Standards | Data Mining | Mixed Metho...
 
Building Ethical AI
Building Ethical AIBuilding Ethical AI
Building Ethical AI
 

More from Mathieu d'Aquin

More from Mathieu d'Aquin (20)

A factorial study of neural network learning from differences for regression
A factorial study of neural network learning from  differences for regressionA factorial study of neural network learning from  differences for regression
A factorial study of neural network learning from differences for regression
 
Recentrer l'intelligence artificielle sur les connaissances
Recentrer l'intelligence artificielle sur les connaissancesRecentrer l'intelligence artificielle sur les connaissances
Recentrer l'intelligence artificielle sur les connaissances
 
Data and Knowledge as Commodities
Data and Knowledge as CommoditiesData and Knowledge as Commodities
Data and Knowledge as Commodities
 
Unsupervised learning approach for identifying sub-genres in music scores
Unsupervised learning approach for identifying sub-genres in music scoresUnsupervised learning approach for identifying sub-genres in music scores
Unsupervised learning approach for identifying sub-genres in music scores
 
Is knowledge engineering still relevant?
Is knowledge engineering still relevant?Is knowledge engineering still relevant?
Is knowledge engineering still relevant?
 
A data view of the data science process
A data view of the data science processA data view of the data science process
A data view of the data science process
 
Dealing with Open Domain Data
Dealing with Open Domain DataDealing with Open Domain Data
Dealing with Open Domain Data
 
Web Analytics for Everyday Learning
Web Analytics for  Everyday LearningWeb Analytics for  Everyday Learning
Web Analytics for Everyday Learning
 
Presentation a in ovive montpellier - 26%2 f06%2f2018 (1)
Presentation a in ovive   montpellier - 26%2 f06%2f2018 (1)Presentation a in ovive   montpellier - 26%2 f06%2f2018 (1)
Presentation a in ovive montpellier - 26%2 f06%2f2018 (1)
 
Learning Analytics: understand learning and support the learner
Learning Analytics: understand learning and support the learnerLearning Analytics: understand learning and support the learner
Learning Analytics: understand learning and support the learner
 
The AFEL Project
The AFEL ProjectThe AFEL Project
The AFEL Project
 
Assessing the Readability of Policy Documents: The Case of Terms of Use of On...
Assessing the Readability of Policy Documents: The Case of Terms of Use of On...Assessing the Readability of Policy Documents: The Case of Terms of Use of On...
Assessing the Readability of Policy Documents: The Case of Terms of Use of On...
 
Data for Learning and Learning with Data
Data for Learning and Learning with DataData for Learning and Learning with Data
Data for Learning and Learning with Data
 
Towards an “Ethics in Design” methodology for AI research projects
Towards an “Ethics in Design” methodology  for AI research projects Towards an “Ethics in Design” methodology  for AI research projects
Towards an “Ethics in Design” methodology for AI research projects
 
AFEL: Towards Measuring Online Activities Contributions to Self-Directed Lear...
AFEL: Towards Measuring Online Activities Contributions to Self-Directed Lear...AFEL: Towards Measuring Online Activities Contributions to Self-Directed Lear...
AFEL: Towards Measuring Online Activities Contributions to Self-Directed Lear...
 
Profiling information sources and services for discovery
Profiling information sources and services for discoveryProfiling information sources and services for discovery
Profiling information sources and services for discovery
 
Analyse de données et de réseaux sociaux pour l’aide à l’apprentissage infor...
Analyse de données et de réseaux sociaux pour  l’aide à l’apprentissage infor...Analyse de données et de réseaux sociaux pour  l’aide à l’apprentissage infor...
Analyse de données et de réseaux sociaux pour l’aide à l’apprentissage infor...
 
From Knowledge Bases to Knowledge Infrastructures for Intelligent Systems
From Knowledge Bases to Knowledge Infrastructures for Intelligent SystemsFrom Knowledge Bases to Knowledge Infrastructures for Intelligent Systems
From Knowledge Bases to Knowledge Infrastructures for Intelligent Systems
 
Data analytics beyond data processing and how it affects Industry 4.0
Data analytics beyond data processing and how it affects Industry 4.0Data analytics beyond data processing and how it affects Industry 4.0
Data analytics beyond data processing and how it affects Industry 4.0
 
Données ouvertes et traces numériques
Données ouvertes et traces numériquesDonnées ouvertes et traces numériques
Données ouvertes et traces numériques
 

Recently uploaded

Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...
Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...
Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...
nirzagarg
 
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
vexqp
 
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
gajnagarg
 
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
nirzagarg
 
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
nirzagarg
 
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
nirzagarg
 
Abortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get CytotecAbortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Riyadh +966572737505 get cytotec
 
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi ArabiaIn Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
ahmedjiabur940
 

Recently uploaded (20)

Introduction to Statistics Presentation.pptx
Introduction to Statistics Presentation.pptxIntroduction to Statistics Presentation.pptx
Introduction to Statistics Presentation.pptx
 
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
TrafficWave Generator Will Instantly drive targeted and engaging traffic back...
 
Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...
Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...
Top profile Call Girls In Bihar Sharif [ 7014168258 ] Call Me For Genuine Mod...
 
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
怎样办理圣地亚哥州立大学毕业证(SDSU毕业证书)成绩单学校原版复制
 
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24  Building Real-Time Pipelines With FLaNKDATA SUMMIT 24  Building Real-Time Pipelines With FLaNK
DATA SUMMIT 24 Building Real-Time Pipelines With FLaNK
 
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
Top profile Call Girls In dimapur [ 7014168258 ] Call Me For Genuine Models W...
 
Vadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book now
Vadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book nowVadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book now
Vadodara 💋 Call Girl 7737669865 Call Girls in Vadodara Escort service book now
 
Nirala Nagar / Cheap Call Girls In Lucknow Phone No 9548273370 Elite Escort S...
Nirala Nagar / Cheap Call Girls In Lucknow Phone No 9548273370 Elite Escort S...Nirala Nagar / Cheap Call Girls In Lucknow Phone No 9548273370 Elite Escort S...
Nirala Nagar / Cheap Call Girls In Lucknow Phone No 9548273370 Elite Escort S...
 
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Tumkur [ 7014168258 ] Call Me For Genuine Models We...
 
Dubai Call Girls Peeing O525547819 Call Girls Dubai
Dubai Call Girls Peeing O525547819 Call Girls DubaiDubai Call Girls Peeing O525547819 Call Girls Dubai
Dubai Call Girls Peeing O525547819 Call Girls Dubai
 
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptxRESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
RESEARCH-FINAL-DEFENSE-PPT-TEMPLATE.pptx
 
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
Top profile Call Girls In Satna [ 7014168258 ] Call Me For Genuine Models We ...
 
Case Study 4 Where the cry of rebellion happen?
Case Study 4 Where the cry of rebellion happen?Case Study 4 Where the cry of rebellion happen?
Case Study 4 Where the cry of rebellion happen?
 
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
Top profile Call Girls In Purnia [ 7014168258 ] Call Me For Genuine Models We...
 
20240412-SmartCityIndex-2024-Full-Report.pdf
20240412-SmartCityIndex-2024-Full-Report.pdf20240412-SmartCityIndex-2024-Full-Report.pdf
20240412-SmartCityIndex-2024-Full-Report.pdf
 
Abortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get CytotecAbortion pills in Jeddah | +966572737505 | Get Cytotec
Abortion pills in Jeddah | +966572737505 | Get Cytotec
 
Digital Transformation Playbook by Graham Ware
Digital Transformation Playbook by Graham WareDigital Transformation Playbook by Graham Ware
Digital Transformation Playbook by Graham Ware
 
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi ArabiaIn Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
In Riyadh ((+919101817206)) Cytotec kit @ Abortion Pills Saudi Arabia
 
Ranking and Scoring Exercises for Research
Ranking and Scoring Exercises for ResearchRanking and Scoring Exercises for Research
Ranking and Scoring Exercises for Research
 
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
Gomti Nagar & best call girls in Lucknow | 9548273370 Independent Escorts & D...
 

Data ethics

  • 1. Data Ethics Mathieu d’Aquin - @mdaquin Data Science Institute National University of Ireland Galway Insight Centre for Data Analytics
  • 3. Data Ethics The set of principles and processes that guide the ethical collection, processing, analysis, use and application of data having an effect on human lives and society d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
  • 4. Data Ethics The set of principles and processes that guide the ethical collection, processing, analysis, use and application of data having an effect on human lives and society d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
  • 5. Data Ethics The set of principles and processes that guide the ethical collection, processing, analysis, use and application of data having an effect on human lives and society d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
  • 6. Data Ethics The set of principles and processes that guide the ethical collection, processing, analysis, use and application of data having an effect on human lives and society d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
  • 7. Ethics What is right, what is fair, what is just. Hosmer, L. T. (1995). "Trust: The Connecting Link between Organizational Theory and Philosophical Ethics". The Academy of Management Review. 20 (2)
  • 8. In an ideal world What is ethical. (right, fair, just) What is legal.
  • 9. In the real world What is ethical. (right, fair, just) What is legal.
  • 10. What does this have to do with data?
  • 11. What is ethical. (right, fair, just) What is legal. What does this have to do with data? Data protection Privacy Statistical bias Black box decisions Uneven access self-governance ...
  • 13. Example related to privacy/data protection In 2014, New York City released data about 173m taxi trips in the city, where the licence plates and identifier of the taxi had been obfuscated for anonymisation purposes. It was de-anonymised within hours of being released… … and later cross-referenced with timestamped pictures of celebrities entering taxis in New York to figure out their personal address, and how much they tipped. See e.g. http://gawker.com/the-public-nyc-taxicab-database-that-accidentally-track-1646724546
  • 14. Example related to privacy/data protection In this case, it is useful to note that: - Replacing identifiers with a hash is not anonymisation, it is at best bad pseudonymisation - Current data protection regulation in Europe regulates against this sort of cases - The upcoming GDPR will make the consequences of this sort of mistakes stronger - It defines its scope as “any information relating to an identified or identifiable natural person ('data subject'); an identifiable natural person is one who can be identified, directly or indirectly”. Arguably, the unanticipated case of the celebrities fall under this scope… and should therefore have been anticipated.
  • 15. But, should also be asking: What is my impact on society? How can I minimise the risk of negative implications? (drawing upon critical social science, and regulation as guidelines) How do I make what I’m doing compliant with regulation? In addition to:
  • 16. Examples related to bias Google search “unprofessional hair for work” and “professional hair for work”
  • 17. Example related to black-box decision The US justice system relies on a tool to predict, when judging for an offence, what is the likeliness an individual has to re-offend. It is based on many variables, including address, type of offence, past history of offences, and ethnicity. It has been demonstrated to make significant mistakes, especially through being prone to give overly negative scores to black people. See https://www.propublica.org/article/machine-bias-risk-assessments-in-criminal-sentencing
  • 18. Notes on those cases - The algorithm is not biased, the data is. Garbage in, garbage out. - Human decisions are not gold standards, and therefore should not be treated as such in training machine learning models - Sometimes, unrelated things just happen to correlate (see http://www.tylervigen.com/spurious-correlations) - a machine learning model will rely on those correlations to make decisions.
  • 19. Should we ban cheese?
  • 20. Example related to uneven access and under-represented cases Researchers at Georgia Institute of Technology developed and used a chatbot to act as a TA for computer science courses (without the students’ knowledge). It worked very well in most cases… … but failed dramatically in uncommon, delicate situation. Bobbie Eicher et al., Jill Watson Doesn’t Care if You’re Pregnant: Grounding AI Ethics in Empirical Studies, AIES 2018
  • 21. Example related to uneven access and under-represented cases Notes on this case: - Another form of bias, not related to spurious or inaccurate correlations, but to under-representation of specific parts of the population. - Raise issues with the uneven access to the benefit of the technology, and therefore unfairness. - “The future is already here — it's just not very evenly distributed” -- William Gibson Bobbie Eicher et al., Jill Watson Doesn’t Care if You’re Pregnant: Grounding AI Ethics in Empirical Studies, AIES 2018
  • 22. Principles for designing ethics data science projects ‘Ethics in Design’ for Data Science Dialectic The process is based on a conversational approach between data and critical social scientists throughout the project’s life-cycle. Reflective Ethical concerns are not pre-fixed; they may emanate from any stage of the project; thus, constant reflexivity on activities and researchers is needed. Creative, not disruptive The objective of this process is to achieve a positive impact on the research, increase its value addressing ethics throughout the project’s life-cycle. All- encompassing Ethical concerns appear as much in the research activities as in their outcomes, their use and exploitation; the process needs to expand on all stages. d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
  • 23. Principles for designing ethics data science projects ‘Ethics in Design’ for Data Science Dialectic The process is based on a conversational approach between data and critical social scientists throughout the project’s life-cycle. Reflective Ethical concerns are not pre-fixed; they may emanate from any stage of the project; thus, constant reflexivity on activities and researchers is needed. Creative, not disruptive The objective of this process is to achieve a positive impact on the research, increase its value addressing ethics throughout the project’s life-cycle. All- encompassing Ethical concerns appear as much in the research activities as in their outcomes, their use and exploitation; the process needs to expand on all stages. d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018 Methodology borrowed from design fiction: the use of speculative and often provocative scenarios involving the artifact to be design (a data process), as a way to explore its possible implications and reflect on their consequences. Pragmatically, it consist in telling stories asking and answering what if questions (e.g. “what if the student is pregnant? What would happen then?”) and building mockups of the final product to reflect on its behaviour. See Anthony Dunne and Fiona Raby, Speculative Everything, MIT Press, 2013 and Joseph Lindley and Paul Coulton, "Back to the Future: 10 Years of Design Fiction". British HCI 2015.
  • 24. Principles for designing ethics data science projects ‘Ethics in Design’ for Data Science Dialectic The process is based on a conversational approach between data and critical social scientists throughout the project’s life-cycle. Reflective Ethical concerns are not pre-fixed; they may emanate from any stage of the project; thus, constant reflexivity on activities and researchers is needed. Creative, not disruptive The objective of this process is to achieve a positive impact on the research, increase its value addressing ethics throughout the project’s life-cycle. All- encompassing Ethical concerns appear as much in the research activities as in their outcomes, their use and exploitation; the process needs to expand on all stages. d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018
  • 25. Principles for designing ethics data science projects ‘Ethics in Design’ for Data Science Dialectic The process is based on a conversational approach between data and critical social scientists throughout the project’s life-cycle. Reflective Ethical concerns are not pre-fixed; they may emanate from any stage of the project; thus, constant reflexivity on activities and researchers is needed. Creative, not disruptive The objective of this process is to achieve a positive impact on the research, increase its value addressing ethics throughout the project’s life-cycle. All- encompassing Ethical concerns appear as much in the research activities as in their outcomes, their use and exploitation; the process needs to expand on all stages. d’Aquin et al, Towards an “Ethics in Design” methodology for AI research projects, in AIES 2018 i.e. don’t do that:
  • 26. Some conclusions Following regulation is insufficient for data ethics. Ethical issues often appear after the development phase, in scenarios that have not been anticipated. Need to uncover those scenarios to integrate in the process ways of mitigating ethical implications, and balance social, economic and ethical values. This cannot be done (currently) by the technologists alone!
  • 27. Shameless self-promotion Check Towards an “Ethics by Design” methodology for AI research projects at the first conference on AI, Ethics and Society, AIES 2018 The Re-Coding Black Mirror worksop at The Web Conference (WWW 2018) - https://kmitd.github.io/recoding-black-mirror/ MagnaCartaForData.org Contacts: mathieu.daquin@insight-centre.ie, mdaquin.net, @mdaquin