In brief A handful of Democrat Senators and House reps say they will introduce legislation that would stop the Feds and cops in the US from using facial-recognition surveillance gear.
The Facial Recognition and Biometric Technology Moratorium Act [PDF] is the latest attempt by lawmakers to regulate the controversial technology. Multiple studies have revealed these algorithms tend to struggle with identifying women and people of color, leading to the software making biased and discriminatory decisions.
A father was wrongfully arrested in Detroit after he was misidentified as a suspected shoplifter by facial-recognition software. The proposed law would prevent the police from relying on this kind of machine-learning surveillance technology to snare people.
“In this moment, the only responsible thing to do is to prohibit government and law enforcement from using these surveillance mechanisms,” said Senator Edward Markey (D-MA) on Thursday. He thanked House reps Pramila Jayapal (D-WA) and Ayanna Pressley (D-MA), and Senator Jeff Merkley (D-OR), for helping him craft and introduce the draft legislation.
As a non-bipartisan bill, the proposed law will need Republican support in the Senate, in particular, as well as the President's signature, to pass. Makers of software and tech for the cops and Feds urged lawmakers to can the proposal, arguing the systems protect Americans.
Amazon's self-driving car spin
Amazon has snapped up self-driving startup Zoox in an attempt to develop its own autonomous driving technology.
The deal was worth a little over $1.2bn, according to the WSJ. That figure was lower than Zoox’s initial $3.2bn evaluation. Now, the upstart will not have to worry about fundraising from investors anymore since it’ll be supported by a trillion-dollar corporation.
Zoox’s CEO Aicha Evans and co-founder and CTO Jesse Levinson will continue to lead the outfit, Amazon said.
Google launches experimental AI image app
Developers working at Google’s experimental Area 120 program have launched an app called Keen that uses AI algorithms to curate basic resources on arbitrary topics that can be shared with others. Here’s what that looks like:
The app pulls together content from Google Search, and uses machine learning to recommend related information. Imagine if a user starts a Keen page for, say, baking bread, they can pull together recipes online or images. That page can then be publicly or privately followed by other people who are also interested in baking bread.
It sounds a little bit like a social media platform. But CJ Adams, co-founder of Keen, said: “[It] isn’t intended to be a place to spend endless hours browsing. Instead, it’s a home for your interests: a place to grow them, share them with loved ones and find things that will help in making this precious life count.”
You can explore Keen here. ®