TFA description doesn't mention the 'Avatar' details at all.
The swarms are cool, but the 'Avatar' program is with *full size* F-15, F-16, and F-18's that are autonomous drones that follow a lead pilot in an F-22 or F-35...from TFA:
One new project not previously reported is called Avatar, and calls for the Pentagon to pair high-tech “fifth-generation” fighter jets like the F-22 Raptor and F-35 Joint Strike Fighter with unmanned versions of older jets like the F-16 Fighting Falcon or F/A-18 Hornet, which would be flown without a pilot for the first time.
The Avatar effort was previously called Skyborg by SCO and is known as “the Loyal Wingman” concept in the Air Force, Roper said. The program will require unmanned fighters to act with enough autonomy that the pilot in the manned jet doesn’t have to direct them all the time.
This is also the future for Google's AI-cars, the actual practical application will be in long-haul trucking. AI will never replace human drivers...Google's cars with no steering wheel will never be implimented. However, we will see the self-driving car tech used in the same way as this aircraft application. One human-driven lead vehicle with AI drones following the human.
Amazon, etc. have so much turnover because their corporate environment is toxic.
They could retain, save all the money on hiring and training, and not have to pay 2x current salary to their employees if they just treated their employees better.
Not only are we asking AI to overcome human error, but we are also asking it to overcome its own. A truly tall order.
well said
i love robotics and AI but the hype is actually hurting what should be a meteoric rise in robotics/AI in all aspects of life. It's happening, but it could be so much better
for my money, the biggest recent advancements in robotics/AI have been in materials science and engineering, not software...we're able to make durable robots smaller and smaller which has been the design barrier more than a coding challenge
to me, the future of robotics/AI is appliances...infinitely less than human, nowhere near sentient AI bots that have a multitude of funcitons
Sphero's BB-8 with the gesture watch is a good example
Virtually all proposals for self driving cars require that the cars be better than the vast majority of human beings before they're set loose on the world
moving goalposts
'better than human' is such a nebulous concept, if $Billions$ are on the line, Google will fudge a bit on "better than human"
throughout history, corporations have behaved this way and any suggestion otherwise is rejected
example: a Google AI car goes way too slow during agressive rush-hour and stops randomly because it is being too 'careful' causing drivers to rear-end the AI Google car...Google puts that as the fault of the "human driver" in their statistics, but from a coding perspective the problem is with Google's AI....however because of how accident liability works, most rear-end situations are legally the fault of the car behind, no matter how irrationally the car in front is driving
it's recorded as "human error" but clearly the problem is with the coding of the AI
this is exactly how Google will do the "as good as a human" paradigm and will cheat it if they have to
from a technical, coding perspective this is absolutely irrational
illiterate humans can do things the best AI researchers in the world cannot begin to program a computer to do: drive across town in aggressive traffic
AI goes slow. A commenter above said they would trade 30 mph MAX for the ability to zone out during car travel...that's a rational opinion at least.
For all the advancements in sensors and geo-spatial awareness, we still cannot begin to program algorithms that make continual life and death decisions based on interpretive cues of other human's probable behavior.
This tech's only practical application is in long haul trucking...it's not as 'cool' or TED-talk worthy as "world of autonomous cars" but it's accurate.
Summers is wrong here, but it is for the same reason he thinks he is right:
"Technology is obviating whatever need there may ever have been for high denomination notes in legal commerce."
Technology is also "obviating" the need for criminals to move large amounts of money in cash at all, and making easier to track them when they do use large sums of cash.
Criminals dealing in $10,000+ cash regularly already have access to several electronic means of transfer and 'laundering' as an alternative.
Technology goes both ways...criminals wouldn't be hampered by removing large denominations because they already have so many electronic options.
women are the secret to success in tech...you want the women in tech who can actually do tech work to come to slashdot first...
*I don't know how* but if the new wave of techie women come to slashdot and find a thriving, positive community they will participate and more importantly promote it to their friends
women in tech want no bullshit news and stuff that matters just as much as the rest of us
slashdot isn't reddit...it's not the worst MRA-nest by far, and i think it's actually a fairly open place, and I don't know how to advise specifically how to to it, but when techie women see a place of respect they will add value to the community like absolutely no other thing in the universe can add
I know absolutely nothing about the company that just bought slashdot, nothing, but judging by your comments on this post you understand the slashdot system and are trying to fix it by tweaking things like firehose weighting...I'm glad you're not trying to re-invent the site.
I've relied on slashdot for *no bullshit* and "see-it-here-first" techie news...what they call "stuff that matters"
More than anything, slashdot for me has been educational. I learn about the issue reading through the comments. Haha, yeah lol, there are trolls and idiots but I just ignored that...the good comments here can be from phd's researching the topic or the engineers who actually code the AI gadget in the article under discussion!
I've been reading since 2001, but didn't even log in to comment until 2006, because I honestly didn't think I had anything to contribute because the level of discussion was so high and relevant. True story!
As long as slashdot has the user-base and maximizes the capabilities of the slashdot CMS to foster productive discussion this will be one of the best techie news sites anywhere!
from an IT and telecommunications perspective it's nothing but network of nodes where the information can be intercepted
the question is not, "Do we want a network inaccessible to criminal or military investigators/spies?"
that's a good question, but it is a moot point...the data *is* accessable by *someone*
how many here on/. really host their own email server in a faraday cage with tor, proxies, etc for all IP traffic with all devices individually encrypted using all the best linux-based secure software?
break one link in the chain and it's done...our digital communications are not secure
now, there is a difference when it comes to a personal device...I don't think a court can force someone to, say, give a password to unlock an iphone...but if the court has proper reason to think it is needed, then I think apple should help the government unlock it
Slashdot Mirror
TFA description doesn't mention the 'Avatar' details at all.
The swarms are cool, but the 'Avatar' program is with *full size* F-15, F-16, and F-18's that are autonomous drones that follow a lead pilot in an F-22 or F-35...from TFA:
This is also the future for Google's AI-cars, the actual practical application will be in long-haul trucking. AI will never replace human drivers...Google's cars with no steering wheel will never be implimented. However, we will see the self-driving car tech used in the same way as this aircraft application. One human-driven lead vehicle with AI drones following the human.
If you extrapolate the growth in processor speed over time, it's clear that in about 20-30 years AI will be unhackable.
Therefore, we shouldn't even bother trying to defend against hacking now and just focus on AI.
You must not know anything about Amazon's corporate culture.
Amazon, etc. have so much turnover because their corporate environment is toxic.
They could retain, save all the money on hiring and training, and not have to pay 2x current salary to their employees if they just treated their employees better.
Also: outsourcing/layoffs
HP, Yahoo, something something
if they think something demonstrably incorrect is correct, they are delusional or ignorant
you said it not me...
AI is not anywhere near capable of what you seem to think
they haven't even scratched the surface of several problems as well...besides the incapable AI, the won't work at all in the rain
You see statements criticizing AI hype **because** this is a tech site for tech people.
If you want air-headed, brainless hype coverage of AI, go to mainstream media.
well said
i love robotics and AI but the hype is actually hurting what should be a meteoric rise in robotics/AI in all aspects of life. It's happening, but it could be so much better
for my money, the biggest recent advancements in robotics/AI have been in materials science and engineering, not software...we're able to make durable robots smaller and smaller which has been the design barrier more than a coding challenge
to me, the future of robotics/AI is appliances...infinitely less than human, nowhere near sentient AI bots that have a multitude of funcitons
Sphero's BB-8 with the gesture watch is a good example
moving goalposts
'better than human' is such a nebulous concept, if $Billions$ are on the line, Google will fudge a bit on "better than human"
throughout history, corporations have behaved this way and any suggestion otherwise is rejected
example: a Google AI car goes way too slow during agressive rush-hour and stops randomly because it is being too 'careful' causing drivers to rear-end the AI Google car...Google puts that as the fault of the "human driver" in their statistics, but from a coding perspective the problem is with Google's AI....however because of how accident liability works, most rear-end situations are legally the fault of the car behind, no matter how irrationally the car in front is driving
it's recorded as "human error" but clearly the problem is with the coding of the AI
this is exactly how Google will do the "as good as a human" paradigm and will cheat it if they have to
no, they cannot
nowhere near this has been done
that point is so ridiculous it doesn't need 'negating' any more than just reading it does
AI cannot eliminate 99% of human errors and there is no rational reason for anyone to think it does.
Your jet had a pilot who was actively monitoring a system and could intervene at any time if the autopilot makes mistakes.
Google and others have made cars designed for no occupant to have any of that control.
yeah, I bet this same 75% of people wouldn't feel safe flying in a jet airliner with no controls or pilot...stupid luddites
YES!
Thank you.
thank you
from a technical, coding perspective AI isn't nearly as advanced or capable as the general consumer thinks
When AI makes the wrong decision it is still "human error".
This whole notion that "AI" can eliminate "human error" is fundamentally erroneous.
HUMANS PROGRAM THE COMPUTER
If an AI's driving algorithm makes the wrong decisions, it's because it was programmed wrong, or was incorrectly tested and evaluated.
"world of autonomous cars"
from a technical, coding perspective this is absolutely irrational
illiterate humans can do things the best AI researchers in the world cannot begin to program a computer to do: drive across town in aggressive traffic
AI goes slow. A commenter above said they would trade 30 mph MAX for the ability to zone out during car travel...that's a rational opinion at least.
For all the advancements in sensors and geo-spatial awareness, we still cannot begin to program algorithms that make continual life and death decisions based on interpretive cues of other human's probable behavior.
This tech's only practical application is in long haul trucking...it's not as 'cool' or TED-talk worthy as "world of autonomous cars" but it's accurate.
Just because they don't think it is safe doesn't mean they are automatically Luddites.
You are wrong to assume their opinions are irrational when you have no way to know.
this is an excellent aspect to the discussion that few people articulate well, thank you
and plus a plan for world domination
internet.org is the non-profit equivalent of a Trojan Horse virus
i'm really proud of India for rejecting this...it's a big win for Net Neutrality
Summers is wrong here, but it is for the same reason he thinks he is right:
Technology is also "obviating" the need for criminals to move large amounts of money in cash at all, and making easier to track them when they do use large sums of cash.
Criminals dealing in $10,000+ cash regularly already have access to several electronic means of transfer and 'laundering' as an alternative.
Technology goes both ways...criminals wouldn't be hampered by removing large denominations because they already have so many electronic options.
hey one last thing,
a piece of advice...
women are the secret to success in tech...you want the women in tech who can actually do tech work to come to slashdot first...
*I don't know how* but if the new wave of techie women come to slashdot and find a thriving, positive community they will participate and more importantly promote it to their friends
women in tech want no bullshit news and stuff that matters just as much as the rest of us
slashdot isn't reddit...it's not the worst MRA-nest by far, and i think it's actually a fairly open place, and I don't know how to advise specifically how to to it, but when techie women see a place of respect they will add value to the community like absolutely no other thing in the universe can add
whipslash, you are doing yeoman's work...
I know absolutely nothing about the company that just bought slashdot, nothing, but judging by your comments on this post you understand the slashdot system and are trying to fix it by tweaking things like firehose weighting...I'm glad you're not trying to re-invent the site.
I've relied on slashdot for *no bullshit* and "see-it-here-first" techie news...what they call "stuff that matters"
More than anything, slashdot for me has been educational. I learn about the issue reading through the comments. Haha, yeah lol, there are trolls and idiots but I just ignored that...the good comments here can be from phd's researching the topic or the engineers who actually code the AI gadget in the article under discussion!
I've been reading since 2001, but didn't even log in to comment until 2006, because I honestly didn't think I had anything to contribute because the level of discussion was so high and relevant. True story!
As long as slashdot has the user-base and maximizes the capabilities of the slashdot CMS to foster productive discussion this will be one of the best techie news sites anywhere!
from an IT and telecommunications perspective it's nothing but network of nodes where the information can be intercepted
the question is not, "Do we want a network inaccessible to criminal or military investigators/spies?"
that's a good question, but it is a moot point...the data *is* accessable by *someone*
how many here on /. really host their own email server in a faraday cage with tor, proxies, etc for all IP traffic with all devices individually encrypted using all the best linux-based secure software?
break one link in the chain and it's done...our digital communications are not secure
now, there is a difference when it comes to a personal device...I don't think a court can force someone to, say, give a password to unlock an iphone...but if the court has proper reason to think it is needed, then I think apple should help the government unlock it
that's different than digital communications
either way it is encrypted
in TFA the car has to be taken off auto-drive because it comes to a construction area and slows down so much it was barely moving
these things are not going to work in this form, and pushing them into the market will be a disaster
car AI is much, much improved, and I can see groups of electric semi-trucks following one lead driver on an interstate, but that's about it
yeah technically it's not multiplexing signals