RE: 5 usi attuali dell'Intelligenza Artificiale che potrebbero ritorcersi contro di noi
I think you've just inspired my next post 🙂
Some of the ideas are quite interesting and these are a few of my random thoughts as I read through...
With the house buying, you mention some kind of "neighbour analysis" and it makes you wonder if the social divide would widen even further as a result of this. There's already evidence that minorities cluster together (as they migrate to another country, often living in close proximity to those they already know) and this could be exacerbated int political groups. And as we see on sites like Facebook and Twitter, this will inevitably lead to tribes or to an extreme, gangs being formed. Which will almost inevitably lead to confrontation. Which happens anyway so maybe this tool will just accelerate things.
The Justice idea just seems completely wrong. Every case would be appealed and I can't see this being adopted in most democratic countries. If it does, it feels like a pointless step before the appeals process begins.
The AI research in medicine is amazing. Given our understanding of many diseases now, it's potential benefits are significant. The biggest barrier that I can think of is the cost of care in most countries. Telling somebody they're at a high risk of something doesn't equate to them getting the treatment that they require.
Argentina won't win the world cup 😜
Thank you, man! Being an inspiration is always nice! 😄
I think that the artificial justice system will be tested at first, at least in Italy, on civil cases, of little value, so as to make any appeal uneconomic. This will artificially increase the reliability rate by declaring sentences such as:
"Did you see? Ninety-five per cent of the cases did not even go to appeal, the system works!"
I still agree with you that every case cannot simply be considered by an algorithm, but must also take into account various 'human' components. Will AI be able to do that?
And above all, will the system be hackable so that the judge's friend, who could previously be saved on a technicality, will not also be punished? This is why I think that AI in justice will never really be fully integrated.
AI in medicine scares me a bit. It's a great opportunity, but what if it expanded to another branch of medicine, such as psychology? For example: according to AI, this subjects, who for example may have been highly critical of the government in the past, is at high risk of violent behaviour, he has to be re-educated at one of our state structures, even if he hasn't committed any offences yet = Minority Report.
All science fiction films announce the future in some way 😯 ?
I don't know who will win the World Cup, but it certainly won't be Italy 😖... It's a tragic year for my football passions!
It'll be interesting to see what types of case they choose to "automate". I can't think of many that can be automated although things like motorists speeding or tax fraud could work well. The idea that lawyers would find a loophole feels inevitable though.
I think there's an element of AI in psychology already and the link to terrorism. Certain behaviours that trigger suspicion and gets people on the "pay close attention to this person" list. I know there are various stats that link somebody who's reported domestic abuse and a subsequent murder (where the abused kills the abuser). This could potentially already be considered AI.
The Simpsons is good at predicting the future...
It's sad that Italy won't be playing but it does have the benefit that you won't beat us again 😄
Yes, I remember the Simpsons scene with Trump on the lift, I think there was also one related to 9/11, the pandemic coming from China and many others. By the way, for The Simpsons the World Cup final will be Mexico Vs Portugal 😧