Good morning, and welcome to a bumper bank holiday newsletter!
Interestingly, there are a few stories on the regulation of AI this week (I wonder why!?). Kicking off with the military side of things, OpinioJuris has a post on responsible AI from the recent REAIM conference. One thing that always intrigues me is why there is not a more specific focus on what human rights law has to offer. The law of armed conflict has evident limitations for the regulation of AI, but it is accepted that human rights law continues to apply, and it can offer a lot in terms of regulating the entire A.I. lifecycle (complementing LoAC). As a side note, much of the discussion on military AI seems to focus on very advanced machines (think the terminator) risking insufficient attention to other uses, particularly in an intelligent context (which will obviously inform all aspects of military operations).
There is a really disturbing report that someone took their life on ChatGPTs recommendation, which is a tragedy of itself, but also highlights the risks of using these types of systems for advice. This is picked up by MedCityNews re healthcare advice.
Quanta magazine has a decent interview on bias, and de-anonymisation.
There are also a few facial recognition stories, following the results of Uk police testing which indicated an absence of bias. Prof Fussey has a thread on potential problems with the study, but beyond algorithmic bias what – I think – is really missing is a discussion on the step change in surveillance capability that facial rec enables (think profiling, tracking, monitoring with minimal police resources) and consideration of the chilling effect of surveillance. What the impact pervasive surveillance might have on society is not a question we should really skip over. (Research underway on both issues)
I’m about to hope on a plane for a really really long time, so I’ll sign off with this really beautiful tune
thanks as ever to Sarah Zarmsky
The Researcher Who Would Teach Machines to Be Fair, Quanta Magazine
ChatGPT is going to change education, not destroy it, MIT Technology Review
Instant Videos Could Represent the Next Leap in A.I. Technology, The New York Times
We are hurtling toward a glitchy, spammy, scammy, AI-powered internet, MIT Technology Review
Council of Europe AI policy framework presented in the Hague, Council of Europe
What About the Robots That Are Already Here? New York City to Begin Enforcement of Artificial Intelligence Applications Related to Applicants and Employees Through the NYC Automated Employment Decision Tools Law on July 5, 2023,Baker Data Counsel
Making Unilateral Norms for Military AI Multilateral, Lawfare Blog
Will AI solve my midlife crisis?, Financial Times
Can We No Longer Believe Anything We See?, The New York Times
Meta won’t say if politicians can post AI-made fakes without warnings, The Washington Post
OpenAI to offer Remedies to Resolve Italy’s ChatGPT Ban, The Washington Post
In A.I. Race, Microsoft and Google Choose Speed Over Caution, The New York Times
Canada sticks a privacy probe into OpenAI’s ChatGPT, The Register
The AI backlash is here. It’s focused on the wrong things., The Washington Post
Can A.I. and Democracy Fix Each Other?, The New York Times
Universities express doubt over tool to detect AI-powered plagiarism, Financial Times
The Opportunities and Risks of AI in Energy Supply, Algorithm Watch
With Google as My Neighbor, Will There Still Be Water?,Algorithm Watch
AI surveillance rumors: gay adult content creators face sanctions, Algorithm Watch
The Ethics of Creating A.I.-Generated Images of Public Figures, Smithsonian Magazine
*Disclaimer: The selected articles and chapters were not evaluated for their research methods and do not necessarily reflect the views of the AI & Human Rights Blog
Defining the scope of AI regulations, Jonas Schuett, Law, Innovation and Technology
Webinar: ‘AI and Gender: Preventing Bias, Promoting Equality’,Council of Europe, 19 April 2023, 1-2pm CET