Dive Brief:
- Last week, the New York City Council passed a bill designed to tackle "algorithmic inequity" in city government and address if tech systems appear to discriminate against people based on a variety of statuses including age, race, religion, gender or sexual orientation. The bill now awaits a signature from Mayor Bill de Blasio.
- The bill summary notes it would "require the creation of a task force that provides recommendations on how information on agency automated decision systems may be shared with the public and how agencies may address instances where people are harmed by agency automated decision systems."
- ProPublica reports that Council Member James Vacca, the bill's sponsor, was inspired to introduce the bill after a May 2016 investigation into machine bias.
Dive Insight:
ProPublica reports that this bill is the first measure of its kind in the country, however New York City is certainly not the first entity to publicly note potential machine bias. According to CIO Dive, social media giants including Facebook, Twitter and Reddit have introduced tools or enacted rules that address hateful content and harassment, which is sometimes left unchecked due to inefficiencies in contextualized decisions made by AI.
"Biases in AI systems can be introduced during the programming itself, as a system learns or during application from a variety of sources. A lack of diversity among the programmers making the system and biased data sets can cause these problems," CIO Dive reports.
Additionally, a recent story of failed iPhone X facial recognition in China sparked the question of diversity in tech talent, and called on tech developers to ensure equal representation of all backgrounds when writing code or developing functionalities. While unrelated to this specific case study, New York's bill presents an opportunity for cities to be proactive about diversity in tech and to design algorithms that are beneficial for all city residents, despite their race, gender, sexual orientation or other status.
If New York's bill is passed into law, it could set a precedent for cities across the U.S. to prioritize appropriate development and monitoring of AI systems — a pressing concern, considering the influx of AI-controlled devices and systems being rolled out nationwide.