[gnur]

Is this about the AI or about the use of regular technology that could hurt citizens? Facial recognition and pattern recognition both seem like tech that isn't necessarily AI but are mentioned in the article.

If it's about AI, where is the line? Does reinforcement learning count? What about deep learning? Neural networks? 5 clever if statements?

[Deukhoofd]

> Social scoring systems, such as those launched in China that track the trustworthiness of people and businesses, are classified as "contravening the Union values" and are going to be banned.

> The proposal also wants to prohibit AI systems that cause harm to people by manipulating their behavior, opinions or decisions; exploit or target people's vulnerabilities; and for mass surveillance.

> many technologies currently in use in Europe today, such as algorithms used to scan CVs, make creditworthiness assessments, hand out social security benefits or asylum and visa applications, or help judges make decisions, would be labeled as "high risk," and would be subject to extra scrutiny.

> One of the Commission’s requirements in the draft are that data sets do not “incorporate any intentional or unintentional biases” which may lead to discrimination.

[patates]

> algorithms used to scan CVs

so "5 clever if statements" (which most CV parsers are, if that wasn't clear) too are subject to extra scrutiny depending on application being labeled high risk or not.

I'm not sure how to feel about this. I feel it's good intentions but marketed wrong.

Why not just say, we will evaluate critical software like we evaluate other critical engineering projects/infrastructure? Why mix in AI?

[AstralStorm]

What does it mean "like other critical projects"?

Do you mean like buildings or lawyers? Certified AI architecture now? Nobody even knows what such process would entail!

The biggest hole is that the AI or statistical systems are not explainable. When push comes to shove, their reasoning is a black hole. Any such system, whether ran by people or automatic, is instant corruption and inequity magnet.

5 if statements are explainable and subject to proper scrutiny.

[navaati]

> 5 if statements are explainable and subject to proper scrutiny.

Except when they are secret. I will never know why my CV got sent to /dev/null that day, was I missing the right keyword for the fourth if ? For me the 5 ifs are as much a blackbox as an advanced neural network.

[deepstack]

> Is this about the AI or about the use of regular technology that could hurt citizens

I would think the intend is about technology that could hurt citizens. More like responsible use of technologies. As a tech I support this kind of initiative. Anything allows software to identify person become a slippery slop to totalitarian government (in this day and age it won't just be government, also include trans-nation corporations).