The Obtain: political AI fashions, and a wrongful arrest

How they did it: The crew requested language fashions the place they stand on numerous subjects, resembling feminism and democracy. They used the solutions to plot them on a political compass, then examined whether or not retraining fashions on much more politically biased coaching knowledge modified their conduct and skill to detect hate speech and misinformation (it did).

Why it issues: As AI language fashions are rolled out into services and products utilized by hundreds of thousands, understanding their underlying political assumptions couldn’t be extra vital. That’s as a result of they’ve the potential to trigger actual hurt. A chatbot providing health-care recommendation may refuse to supply recommendation on abortion or contraception, for instance. Learn the total story. 

—Melissa Heikkilä

Learn subsequent: AI language fashions have not too long ago turn out to be combined up within the US tradition wars, with some calling for builders to create unbiased, purely fact-based AI chatbots. In her weekly e-newsletter all about AI, The Algorithm, Melissa delves into why it’s a pleasant idea—however technically not possible to construct. Learn it to seek out out extra, and if you happen to don’t already, signal as much as obtain it in your inbox each Monday. 

The must-reads

I’ve combed the web to seek out you right this moment’s most enjoyable/vital/scary/fascinating tales about expertise.

1 A lady was wrongfully arrested after a false face recognition match
It’s notable that each individual we all know this has occurred to has been Black. (NYT $)
+ The motion to restrict face recognition tech may lastly get a win. (MIT Expertise Evaluate)
 
2 AI startups are combating soiled 😈
We’re speaking pretend names, rivals posing as prospects, and even bombing Zoom calls. (NYT $)
+ It’s all beginning to look quite a bit like a bubble. (WP $) 
 
3 A vote in San Francisco may change the way forward for driverless vehicles 
All eyes are on whether or not the state board will approve an enormous growth of autonomous taxis on Thursday. (NBC)
+ Massive tech firms are struggling to win over native residents and public officers. (WSJ $)
 
4 Is Texas’ electrical energy grid going to have the ability to deal with electrical automobiles?
There are causes to be optimistic, not only for that state however the US as an entire. (The Atlantic $)
 
5 Criminals are enthusiastic early adopters of AI instruments
On the darkish net, they declare to have created two massive language fashions that may help with unlawful actions. (Wired $)
+ Criminals are additionally utilizing AI-generated books to rip-off folks. (NYT $)
+ We’re hurtling towards a glitchy, spammy, scammy, AI-powered web. (MIT Expertise Evaluate)
 
6 The period of plentiful low-cost stuff could also be coming to an finish 
Possibly that’s not a completely unhealthy factor, frankly, for the sake of the planet. (WSJ $)
 
7 Persons are eager to recreate Black Twitter elsewhere
There’s been an enormous exodus from the positioning. However the place ought to of us go? (WP $) 
 
8 Massive cities want to alter
To thrive, they should reinvent themselves to be extra than simply locations the place folks work. (Vox)
+ What cities want now. (MIT Expertise Evaluate)
 
9 WhatsApp is engaged on 32-person voice chats
Seems like pure chaos! (The Verge) 
 
10 Even Zoom is making staff return into the workplace
Ironic, maybe. However not that shocking. (Quartz $)

Quote of the day