The legal framework is one of the first laws globally to regulate the use of artificial intelligence for “high risk” applications like hiring and policing. /2
They will evaluate AI-applications based on risk:
https://abs.twimg.com/emoji/v2/... draggable="false" alt="⚫" title="Schwarzer Kreis" aria-label="Emoji: Schwarzer Kreis"> Unacceptable
https://abs.twimg.com/emoji/v2/... draggable="false" alt="🔴" title="Roter Kreis" aria-label="Emoji: Roter Kreis"> High risk
https://abs.twimg.com/emoji/v2/... draggable="false" alt="🟠" title="Orangefarbener Kreis" aria-label="Emoji: Orangefarbener Kreis"> Limited risk
https://abs.twimg.com/emoji/v2/... draggable="false" alt="🟢" title="Grüner Kreis" aria-label="Emoji: Grüner Kreis"> Minimal risk
https://ec.europa.eu/commission/presscorner/detail/en/QANDA_21_1683
/3">https://ec.europa.eu/commissio...
They will evaluate AI-applications based on risk: https://abs.twimg.com/emoji/v2/... draggable= Unacceptablehttps://abs.twimg.com/emoji/v2/... draggable="false" alt="🔴" title="Roter Kreis" aria-label="Emoji: Roter Kreis"> High riskhttps://abs.twimg.com/emoji/v2/... draggable="false" alt="🟠" title="Orangefarbener Kreis" aria-label="Emoji: Orangefarbener Kreis"> Limited riskhttps://abs.twimg.com/emoji/v2/... draggable="false" alt="🟢" title="Grüner Kreis" aria-label="Emoji: Grüner Kreis"> Minimal risk https://ec.europa.eu/commissio..." title="They will evaluate AI-applications based on risk: https://abs.twimg.com/emoji/v2/... draggable="false" alt="⚫" title="Schwarzer Kreis" aria-label="Emoji: Schwarzer Kreis"> Unacceptablehttps://abs.twimg.com/emoji/v2/... draggable="false" alt="🔴" title="Roter Kreis" aria-label="Emoji: Roter Kreis"> High riskhttps://abs.twimg.com/emoji/v2/... draggable="false" alt="🟠" title="Orangefarbener Kreis" aria-label="Emoji: Orangefarbener Kreis"> Limited riskhttps://abs.twimg.com/emoji/v2/... draggable="false" alt="🟢" title="Grüner Kreis" aria-label="Emoji: Grüner Kreis"> Minimal risk https://ec.europa.eu/commissio..." class="img-responsive" style="max-width:100%;"/>
Some technologies are banned or "Unacceptable": A surveillance ban on AI systems that track people indiscriminately – f.ex. some restrictions on biometric identification ("we’re not China!") /4
Also, a ban on social credit scores that track individual behavior, impact hiring and judiciary decisions, and rate trustworthiness – f. ex. ban on social scoring (good) /5
AI systems for military purposes are outside the scope of this risk evaluation, like was expected. /6
Apart from the proposal for a legal framework, they launched a "A Coordinated Plan with EU countries" and "New rules on machinery products". They want to invest €1 billion per year in AI technologies. /7
The EU wants to avoid the worst of what artificial intelligence can do while still trying to boost its potential for the economy in general - But how will (tech) companies respond to these regulations? /9
Also, in the draft document, there is a cacophony of definitions of AI (more than 40!) - The concept and the central technologies itself need to be more elucidated /10
In conclusion, this legal framework is still a proposal, the three main EU bodies (Council, Parliament and Commission) will have to agree on a common version of the text. To be continued... /11
Other interesting thread, more from the legal/data protection lens: https://twitter.com/gabrielazanfir/status/1384821003893760003">https://twitter.com/gabrielaz...
You can follow @CoolsHannes.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: