ケンブリッジ大学主催「Alの未来」ワークショップに参加しました。

2019年3月25日、ケンブリッジ大学主催のICTのAl、AICT(=AI+ICT)に関する「Alの未来」ワークショップに参加しました。

英国議会の人工知能委員会、ケンブリッジ大学、ロンドン大学、グーグルのエンジニアなどが、ディープラーニングの限界やデータのバイアス、倫理概念の定義の難しさなど、様々な課題について紹介しました。

例えば、ケンブリッジ大学のエイドリアン・ウェラー(Adrian Weller)博士は、「透明性」の概念一つとっても、開発者、利用者、規制など政策立案者の立場によって意味するものが違うため、定義つけることは困難であると述べています。

このように、AI倫理で用いられている概念は(「公平性」や「説明責任」、「透明性」など)非常に高度で抽象度が高い一方で、現実社会における倫理は個々の社会的文脈に依拠しているのです。

The Future of Artificial Intelligence: Language, Ethics, Technology
25 March 2019, 10:00 – 17:00
Room SG1, The Alison Richard Building, 7 West Road, Cambridge, CB3 9DT

This is the inaugural workshop of Giving Voice to Digital Democracies: The Social Impact of Artificially Intelligent Communications Technology, a research project which is part of the Centre for the Humanities and Social Change, Cambridge and funded by the Humanities and Social Change International Foundation.​

The workshop will bring together experts from politics, industry, and academia to consider the social impact of Artificially Intelligent Communications Technology (AICT). The talks and discussions will focus on different aspects of the complex relationships between language, ethics, and technology. These issues are of particular relevance in an age when we talk to Virtual Personal Assistants such as Siri, Cortana, and Alexa ever more frequently, when the automated detection of offensive language is bringing free speech and censorship into direct conflict, and when there are serious ethical concerns about the social biases present in the training data used to build influential AICT systems.

Speakers

Professor Emily M. Bender, University of Washington
Baroness Grender MBE, House of Lords Select Committee on AI
Dr Margaret Mitchell, Google
Dr Melanie Smallman, UCL, Alan Turing Institute
Dr Marcus Tomalin, University of Cambridge
Dr Adrian Weller, University of Cambridge, Alan Turing Institute, The Centre for Data Ethics and Innovation

Giving Voice to Digital Democracies explores the social impact of Artificially Intelligent Communications Technology – that is, AI systems that use speech recognition, speech synthesis, dialogue modelling, machine translation, natural language processing, and/or smart telecommunications as interfaces. Due to recent advances in machine learning, these technologies are already rapidly transforming our modern digital democracies. While they can certainly have a positive impact on society (e.g. by promoting free speech and political engagement), they also offer opportunities for distortion and deception. Unbalanced data sets can reinforce problematical social biases; automated Twitter bots can drastically increase the spread of malinformation and hate speech online; and the responses of automated Virtual Personal Assistants during conversations about sensitive topics (e.g. suicidal tendencies, religion, sexual identity) can have serious consequences.

Responding to these increasingly urgent concerns, this project brings together experts from linguistics, philosophy, speech technology, computer science, psychology, sociology, and political theory to develop design objectives for the creation of AICT systems that are more ethical, trustworthy, and transparent. These technologies will have the potential to affect more positively the kinds of social change that will shape modern digital democracies in the immediate future.

Program
9.30 – 10.00
Registration

10.00 – 10.30
Marcus Tomalin (University of Cambridge)

Welcome and Introduction

10.30 – 11.15
Baroness Grender MBE (House of Lords Select Committee on AI)

‘AI Ready, Willing and Able? What Can the Government Do?’

11.15 – 11.30
Break

11.30 – 12.15
Melanie Smallman (University College London/Alan Turing Institute)

‘Fair, Diverse and Equitable Technologies: The Need for Multiscale-Ethics’

12.15 – 13.00
Adrian Weller (University of Cambridge/Alan Turing Institute/The Centre for Data Ethics and Innovation)

‘Can We Trust AI Systems?’

13.00 – 14.00
Lunch

14.00 – 14.45
Marcus Tomalin (University of Cambridge)

‘The Ethics of Language and Algorithmic Decision-making’

14.45 – 15.30
Emily M. Bender (University of Washington)

‘A Typology of Ethical Risks in Language Technology with an Eye Towards Where Transparent Documentation Can Help’

15.30 – 15.45
Break

15.45 – 16.30
Margaret Mitchell (Google)

‘Bias in the Vision and Language of Artificial Intelligence’

16.30 – 17.00
Round Table Discussion

 

 

About Toshie Takahashi

Toshie Takahashi is Professor in the School of Culture, Media and Society, Waseda University, Tokyo and Faculty Associate at the Harvard Berkman Klein Center for Internet & Society. She has held visiting appointments at the University of Oxford and the the University of Cambridge as well as Columbia University. She conducts cross-cultural and trans-disciplinary research on the social impact of robots as well as the potential of AI for Social Good. 【早稲田大学文学学術院教授。ハーバード大学バークマンクライン研究所ファカルティ・アソシエイト。ケンブリッジ大学ならびにコロンビア大学客員研究員。現在、人工知能の社会的インパクトやロボットの利活用などについて、ハーバード大学やコロンビア大学、ケンブリッジ大学と国際共同研究を行っている。東京オリンピック・パラリンピック競技大会組織委員会テクノロジー諮問委員会委員。】
This entry was posted in AI, グローバリゼーション, テクノロジー, 第4次産業革命. Bookmark the permalink.