In preparing this new website, I made some choices about what work to highlight, and what aspects of my career to emphasize in the future. I therefore have a pretty accurate idea of my work, my publications and its impact. I also decided to ask Chat GPT — the artificial intelligence chatbot that is claimed to be driving the next era of digital innovation. Much has already been written the potential non-compliance with global and Canadian privacy standards. Data protection authorities, including the Office of the Privacy Commissioner of Canada, have announced investigations.
The principal concern, it seems, is the scraping of personal data without the knowledge and consent of the individual to “train” the “large language model” (LLM) that forms the basis of this generative artificial intelligence system. Presumably some of those data are in sensitive “special categories” defined by the General Data Protection Regulation (GDPR). Some may relate to children, a concern that prompted the Italian DPA to ban the use of Chat GPT in Italy.
A related concern is the inability of individuals to exercise their privacy rights over their data, and in particular to correct and/or delete data that might be incorrect, obsolete or incomplete. In broader terms, the European courts have recognized a “right to be forgotten” to that reputations can be protected. There is considerable confusion about how, and if, that right can be exercised in relation to generative AI systems.
Like many other people, I was curious what Chat GPT would spew out if I asked it about myself. Here is the record of the conversation.
Who is Colin Bennett? What does he research and write about? What are his most significant publications?
Colin Bennett is a Canadian political scientist and professor at the University of Victoria in British Columbia. He is known for his research on the politics of privacy and surveillance, as well as information technology and public policy.
Bennett has written extensively on the subject of privacy and surveillance, exploring the impact of technological advancements on these issues. His work focuses on the ways in which governments, corporations, and other organizations collect and use personal information, and the implications this has for individual privacy and civil liberties.
(So far so good)
Some of his most significant publications include:
- The Privacy Advocates: Resisting the Spread of Surveillance (2008)
- The Governance of Privacy: Policy Instruments in Global Perspective (2003)
- Regulating Privacy: Data Protection and Public Policy in Europe and the United States (1992)
- The Electronic Eye: The Rise of Surveillance Society (1994)
(Three out of four)
Bennett did not write The Electronic Eye. That book is by David Lyon. Please suggest other publications
I apologize for the error in my previous response. You are correct that “The Electronic Eye” was written by David Lyon, not Colin Bennett.
Here are some of Colin Bennett’s other significant publications:
- The Privacy Advocates: Resisting the Spread of Surveillance (2008)
- The Governance of Privacy: Policy Instruments in Global Perspective (2003)
- Regulating Privacy: Data Protection and Public Policy in Europe and North America (1992)
- Visions of Privacy: Policy Choices for the Digital Age (1999)
- Transparent Lives: Surveillance in Canada (2014)
- Privacy, Surveillance and the New ICTs (2016)
(5 out of 6)
I do not recognize “Privacy, Surveillance and the new ICTs.” Where was this published?
I apologize for the confusion. “Privacy, Surveillance and the New ICTs” is actually a book chapter written by Colin Bennett that appears in the edited volume “The Routledge Handbook of Surveillance Studies” (2016), which was edited by Kirstie Ball, Kevin Haggerty, and David Lyon.
In the chapter, Bennett examines the impact of new information and communication technologies (ICTs) on privacy and surveillance practices, and explores the regulatory challenges that arise as a result. The chapter provides a comprehensive overview of the ways in which new ICTs are transforming surveillance practices, and offers insights into how policymakers can address these changes while protecting privacy and civil liberties.
No that is wrong too. This handbook was published in 2012, and his chapter was called “Privacy Advocates, privacy advocacy and the surveillance society.”
I apologize for the mistake again. You are correct that “The Routledge Handbook of Surveillance Studies” was first published in 2012, and Colin Bennett’s chapter in the book was titled “Privacy Advocates, Privacy Advocacy and the Surveillance Society.”
In the chapter, Bennett discusses the emergence of privacy advocacy as a response to growing surveillance practices, and examines the role of civil society organizations in promoting privacy rights and protecting against surveillance. The chapter provides an overview of different privacy advocacy strategies and tactics, and explores the challenges and opportunities faced by privacy advocates in the face of technological advancements and changing political landscapes.
Thank you for bringing this to my attention, and I apologize for any confusion or misinformation caused by my previous responses.
I then tried to test whether this correction was, indeed, genuine.
In the article, Bennett discusses the impact of new information and communication technologies (ICTs) on privacy and surveillance practices, and explores the regulatory challenges that arise as a result. He argues that the emergence of new ICTs has fundamentally transformed the nature of surveillance, and that traditional regulatory approaches are inadequate to address these changes.
Bennett calls for a rethinking of privacy and surveillance regulation in the digital age, and proposes a new framework for governing privacy and surveillance practices. He emphasizes the importance of transparency, accountability, and citizen participation in the development of new privacy and surveillance policies, and argues that effective privacy protection requires a holistic approach that incorporates legal, technological, and social strategies.
I have never written an article with this title, and I have never published in the journal Current Sociology. And as far as I can determine, nobody named Bennett published any article with this banal title. Therefore, this inane description is totally fabricated. It is clear that this “conversation” could go on ad infinitum with a phony “apology” before every response.
So what is going on here? Am I actually exercising any rights here? Am I using Chat GPT to actually correct the personal data that the model generates and disseminates about me, and which is indeed my right under the Personal Information Protection and Electronic Documents Act (PIPEDA)? Open AI is clearly processing personal data in Canada to train its language model. It clearly has to comply with this legislation.
Furthermore, OPEN AI’s privacy policy is explicit that:
“Depending on location, individuals in the EEA, the UK, and across the globe may have certain statutory rights in relation to their Personal Information. For example, you may have the right to:
- Access your Personal Information.
- Delete your Personal Information.
- Correct or update your Personal Information.
- Transfer your Personal Information elsewhere.
- Withdraw your consent to the processing of your Personal Information where we rely on consent as the legal basis for processing.
- Object to or restrict the processing of your Personal Information where we rely on legitimate interests as the legal basis for processing.”
On one level, this is a trivial issue of scholarly authorship which matters little to anybody except myself or perhaps others who might want to cite me. My invitation to Chat GPT to reassess the fictitious final article did not lead to a checking of its veracity against any third party source (including my own personal website). And that is what typically occurs when rights of correction or rectification are exercised. In Chat GPT, the process of “rectification” is constant, iterative and endlessly manipulable. And that process is inherently dangerous when information is generated about individuals that can really harm reputation.
I suppose I should be gratified that Chat GPT concluded that:
Bennett’s work on privacy and surveillance has been influential in shaping public policy debates and academic scholarship on these issues.
And I guess that is my hope in this post! The intersection of established rights to rectification and the operation of these LLMs is just one of several aspects of privacy compliance that Canadian and international regulators will need to investigate.