The UK government had its own AI built to block extremist video
New software, intended for use by small tech firms, can pick out ISIS propaganda with reasonable accuracy.
The news: Developed by London-based data science firm ASI on behalf of the British government, the $800,000 AI is trained on thousands of hours of video to spot extremist content. It can identify 94 percent of ISIS propaganda with 99.99 percent accuracy, and is designed to be used to block the upload of such material.
Who it's for: It will be offered to small tech firms that can’t afford to develop such systems. The likes of Facebook and YouTube already have similar algorithms, though the British government has been critical of how widely and quickly they have been put to use by tech giants.
AI by law? The UK’s home secretary tells the BBC that use of the new AI could potentially become a legal requirement.
But: Because the AI is trained on historical data, extremists will likely be able to develop new ways to circumvent its watchful eye.
Deep Dive
Tech policy
How conservative Facebook groups are changing what books children read in school
Parents are gathering online to review books and lobby schools to ban them, often on the basis of sexual content.
Why can’t tech fix its gender problem?
A new generation of tech activists, organizers, and whistleblowers, most of whom are female, non-white, gender-diverse, or queer, may finally bring change.
How the idea of a “transgender contagion” went viral—and caused untold harm
A single paper on the notion that gender dysphoria can spread among young people helped galvanize an anti-trans movement.
The world is moving closer to a new cold war fought with authoritarian tech
At the Shanghai Cooperation Organization summit, Iran, Turkey, and Myanmar promised tighter trade relationships with Russia and China.
Stay connected
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.