News briefing 13-17 May 2024

Autonomy in weapons systems, and AI in security and defence

• International: The Machine Got it Wrong? Uncertainties, Assumptions, and Biases in Military AI: Arthur Holland Michel, a journalist and researcher writes for Just Security on emerging technologies. Michel puts it simply and says ‘systems can be biased’ and more importantly that ‘Governments often talk about biases in AI systems, but they never talk about uncertainties and assumptions. He emphasizes that States ‘need to be clear that lessons from the civilian uses of AI will only get you so far in battle.’
• International: The President’s Inbox Recap: The Impact of AI on Warfare: The President’s Inbox is a podcast by the Council for Foreign Relations, which features a conversation between Andrew Reddie, an associate research professor of public policy at the University of California Berkeley and the host James M. Lindsay in this episode. In this conversation on the impact of AI on warfare, Reddie warns against the conflation between military AI and autonomous weapons and notes that decision support systems raise questions on responsibility and accountability. 
• International: Tech workers should shine a light on the industry’s secretive work with the military: This piece by William Fitzgerald in MIT Technology Review outlines the nexus between private tech companies and the military in producing military AI. He calls for a sustained movement by tech workers to ensure that tech companies stop profiteering from its contribution in ‘developing the deadliest weapons imaginable.’ 

AI, algorithms and autonomy in the wider world

• US/China: US Lawmakers seek $32 billion to keep American AI ahead of China: An interesting piece this week covered from Reuters covers a report on ‘concerns over China’s “misuse” of AI, at a time when China has engaged on the ‘first formal bilateral talks with the US’ and is said to ‘to open a new front in its effort to safeguard U.S. AI from China and Russia.’

Regulation and Enforcement

• International: Autonomous weapons: IPU adds parliamentary weight to pressure for regulation: This piece by the Inter- Parliamentary Union highlights the growing global momentum for regulation of autonomous weapon systems, and the likely impact on this of the IPU resolution ‘Addressing the social and humanitarian impact of autonomous weapon systems and artificial intelligence.’ The piece further says that ‘by adding these issues to its agenda, the IPU has opened a new forum for multilateral discussion and added parliamentary weight to the growing demand for regulation of AWS and AI at the international level.’ 

Facial recognition, biometric identification, surveillance

• US: Clearview AI facial recognition gets green light for use by Dallas police: Biometric Update reports that the city council of Dallas, Texas ‘has approved the use of Clearview AI’s algorithmic 1-n biometric face matching software for law enforcement purposes. The system uses a face-scraping process that sends potential biometric matches found on the internet for review by two human officers.’ 
• US: Supplemental Written Testimony on Civil Rights Implications of Federal Use of Facial Recognition Technology: Latino civil rights organization in US, UnidosUS has submitted a supplemental written testimony to the US Commission on Civil Rights regarding the use of facial recognition softwares by the Department of Homeland Security. The submission has noted that while broad ethical guidelines for the ethical use of these softwares have been released, concrete and enforceable regulation and protections are missing. 
• New Zealand: Discrimination concerns arise as NZ police expand biometric data use: This piece by Biometric Update reports on an upgrade by the New Zealand Police to their ‘automated fingerprint identification system (AFIS) and biometric recording devices, aiming to modernize and enhance their capabilities in crime investigation and evidence collection.’ This move has raised fresh concerns of ‘bias and discrimination, particularly against Māori and Pasifika populations.’ 
• Tanzania: Court orders changes to Tanzania’s data protection law: Biometric Update reports that the High Court of Tanzania has directed the country’s legislature to amend sections of the Personal Data Protection Act (PDPA) 2022 . These amendments are to be aimed at providing more clarity on provisions which mention ‘unlawful means of procuring and processing personal data.’