Google Launches Global Council To Advise on AI and Tech Ethics (reuters.com) 62
Google announced today that it has formed an external advisory group -- dubbed the Advanced Technology External Advisory Council (ATEAC) -- that is tasked with "considering some of the most complex challenges in AI," including facial recognition and fairness in machine learning. From a report: The council, which is slated to publish a report at the end of 2019, includes technology experts, digital ethicists, and people with public policy backgrounds, Kent Walker, Google's senior vice president for global affairs, said at a Massachusetts Institute of Technology conference. The group is meant to provide recommendations for Google and other companies and researchers working in areas such as facial recognition software, a form of automation that has prompted concerns about racial bias and other limitations. "We want to have the most informed and thoughtful conversations we can," Walker said on stage at the MIT Technology Review event in San Francisco. "We want to sit down with the council and see what agenda they want to set."
Re: (Score:2)
Thanks but no thanks (Score:3, Insightful)
the truth about wisdom (Score:2)
The other day I figured out what "wisdom" really is: it's when your counterfactuals become a lot less shiny.
Many of the youthful believe that if Trump hadn't been elected, America would be a shiny, perfect place. (They won't say this out loud, of course, being generation cynical.) Or they believe that if Trump were impeached, America would return to its former glory as a shiny, perfect place.
I personally kind of hope Trump gets reelected in 2020 (along with a Democrat house and a Democrat senate) so that he
(((Powers That Be))) (Score:1, Insightful)
It's a typical trick that works every time: Set up foundations, councils, duly elected members, special iconography, annual awards ceremonies, and special titles of office. Then, nobody questions your actions; they're obviously official and well thought out.
"Google receives the 2025 Azimov Award for Ethical AI!" will read the headlines.
It's all a show. (((They))) know how to put on a good show.
to whom? (Score:2)
Are they going to advise "ethics" to their Chinese state customers, or just maunder on about how guilty white men in tech should feel about not having a vagina?
I know, let's discuss the ethics of hiring H1Bs from overseas instead of local people?
Facial recognition, fairness in machine learning (Score:1)
Nations all over the world use working facial recognition on all their citizens and tourists everyday with the results expected for the cost of the systems.
Want great facial recognition? Look around d the world and buy into great facial recognition tech.
Machine learning is going to take in criminal, police and CCTV data. It will show who is doing crime and in what in
Re: (Score:2)
I'm pretty sure whoever modded you down failed to notice a 'whoosh' sound overhead. But to be fair to that moderator, you post does ramble a bit, and it took me a while to be only fairly certain that you're being sarcastic. If it hadn't been for your username and your sig I might not have clued in to that possibility.
Re: (Score:1)
Want great facial recognition tech that works over an entire population, invest in it until it always works.
Machine learning is going to get its data sets from criminal populations all over the USA.
Such data sets will reflect what the FBI found over decades about inner city crime and criminals.
Give machine learning data sets about a nations recidivism rate and see what the data shows.
CCTV is mostly ready with gait, automatic number plate recognition. Facial recognition is
Re: (Score:1)
Lots of criminals get found.
Illegal immigrants find their state issued sanctuary city ID won't work in a city/state/federally.
Why should witness protection allow for all the crimes and illegal immigrants?
The new ID given by witness protection will be full working US citizens ID with all needed database history.
Re "our spies"? US spies should be working outside the USA and be provided with a working ID and story as to their past/history/education/work/faith/social m
Re: Facial recognition, fairness in machine learni (Score:2)
The cops in most big American cities can't be bothered to stop people committing crime right outside the police station. The ONLY thing panoptic surveillance will be used for is repressing political dissent.
Cybernetic totalitarianism is already here. The Machine is always watching.
Re: (Score:1)
The new AI, machine learning should track that inner city crime rate.
The way to make it working
Re: Facial recognition, fairness in machine learn (Score:2)
The major problem is that the police can't be bothered to fight actual crime. This will not be improved by fitting them out with the equipment of totalitarianism.
Re: (Score:1)
Inner city crime rates.
What gentrification results in.
What new rental assistance does in once low crime middle and upper class areas.
Re: Facial recognition, fairness in machine learn (Score:2)
Who knows, it might help a little with crime. After all they say East Germany was pretty safe...
Wait a minute, who are we kidding? These are American cops. We're going to get totalitarian repression AND it will still be unsafe to walk downtown at night.
Re: (Score:2)
Ioanid Gang https://en.wikipedia.org/wiki/... [wikipedia.org]
A 1960 propaganda film had to be made to show how they got tracked down.