What are the current limits of artificial intelligence? & their weaknesses

After the worldwide hack of last days I start to wonder what will happen when this would be used to attack the financial markets? AI can be life threatening for the whole world. What would happen if the hack would start a unstoppable crash? Or are we sure we can keep everything under control?

I am afraid we cannot.

It might be the modern version of this:

On June 18th, 74,000 French troops led by Napoleon, sizing up to meet 67,000 British and other European Troops 200 miles NE of Paris.

Nathan Rothschild knowing that information is power stationed his trusted agent named Rothworth near the battlefield. As soon as the battle was over Rothworth quickly returned to London, delivering the news to Rothschild 24 hours ahead of Wellington's courier. A victory by Napoleon would have devastated Britain's financial system. Nathan stationed himself in his usual place next to an ancient pillar in the stock market. Knowing he would be observed he hung his head and began openly to sell huge numbers of British Government Bonds. Believing this to mean that Napoleon must have won, everyone started to sell their British Bonds as well. The bottom fell out of the market. Rothschild had his agents buying up all the hugely devalued bonds.
 
Last edited:
Mtrader said:
After the worldwide hack of last days I start to wonder what will happen when this would be used to attack the financial markets? AI can be life threatening for the whole world. What would happen if the hack would start a unstoppable crash? Or are we sure we can keep everything under control?

I am afraid we cannot.

It might be the modern version of this:

On June 18th, 74,000 French troops led by Napoleon, sizing up to meet 67,000 British and other European Troops 200 miles NE of Paris.

Nathan Rothschild knowing that information is power stationed his trusted agent named Rothworth near the battlefield. As soon as the battle was over Rothworth quickly returned to London, delivering the news to Rothschild 24 hours ahead of Wellington's courier. A victory by Napoleon would have devastated Britain's financial system. Nathan stationed himself in his usual place next to an ancient pillar in the stock market. Knowing he would be observed he hung his head and began openly to sell huge numbers of British Government Bonds. Believing this to mean that Napoleon must have won, everyone started to sell their British Bonds as well. The bottom fell out of the market. Rothschild had his agents buying up all the hugely devalued bonds.
More...

If the powers that be decides to, they can irreversably devastate the world in 30 minutes. We live when the rational choice is to not destroy the world, and because of nothing else.

As for modern markets, the more "sophisticated" they become, the more circuit breakers they require at all scales. Food for thought.
 
GREAT. Both the US, UK, China and Russia are going to spend more on their militaries.
Getting secure systems in place might be a better investment. There are already enough bombs to wipe out life on planet earth.
 
Humpy said:
There are already enough bombs to wipe out life on planet earth.
More...

I wonder if a the computersystem all over the country would crash, and eventually internet will crash too,will it still be possible to launch any bombs?

Or are these bombs launched manually? :D
 
One of too many times being too close to nuclear war:

"While being depth-charged to the surface, the submarine’s captain and the political officer voted to fire their [nuclear] torpedo at a United States Navy target. The third officer, Second Captain Vasili A. Arkhipov, voted nay.

This cautionary tale teaches many lessons. What a president knows, or thinks he knows, in a severe crisis can be less important than what he doesn’t know. Hasty decision making can have catastrophic consequences, while diplomacy can be a far safer course than pre-emptive strikes."

https://www.nytimes.com/2017/04/13/opinion/how-to-make-the-nuclear-button-safer.html?_r=0

Now we can only speculate based on rational logic and historical facts, but I think it's safe to assume that the process of "nuclear button" will involve few persons, hardened or no technology, encryption/keywords, but not something which can't be decided remotely within 10-15 minutes, maybe less, but which also must require human participation and local acknowledgement. The development of new nuclear offences and defenses constantly challenge the old and simple designs, but as long as there is commitment not to automate to 100% efficiency and not to skew the balance too much, humans as species have a fighting chance.

There's probably no involvement of politicians required for MAD responses, and there are already several historical incidents where only one human (usually military personell of lower ranks) prevented the ultimate response, some known, and maybe more unknown. When a MAD response is implemented, the scale of extinction will be down to microbes and maybe some small hardened insects. This is why shelters were abandoned early on in the 60's as an idea of protection, as the nuclear winter after hydrogen bombs could be too long and hard to survive even in shelters, not much life is suspected to be left in the aftermath.

MAD ensures that once a nuclear war is on, it may escalate quickly and uncontrolled, but that's not necessary for extinction, just to make sure the response was met in time. So a MAD war probably means extinction many times over, as responses meets responses all over the world.

Although debatable, better to be on the safer side?
https://www.nytimes.com/2016/02/11/opinion/lets-end-the-peril-of-a-nuclear-winter.html
 
Last edited:
ironchef said:
I don't think you need to worry about the US, as we still use very "secured" technology:

http://www.npr.org/sections/thetwo-...s-on-outdated-technology-such-as-floppy-disks
More...

"However, to address obsolescence concerns, the floppy drives are scheduled to be replaced with Secure Digital devices by the end of 2017. Modernization across the entire Nuclear Command, Control, and Communications (NC3) enterprise remains ongoing."​

The problem is, these "Secure Digital Devices" simply don't exist. USB? USB-C? They're much, much worse than floppy-disks security-wise actually!

Maybe, by using quantum encryption across a secure channel, but for what purposes and distances?
 
Simples said:
"However, to address obsolescence concerns, the floppy drives are scheduled to be replaced with Secure Digital devices by the end of 2017. Modernization across the entire Nuclear Command, Control, and Communications (NC3) enterprise remains ongoing."​

The problem is, these "Secure Digital Devices" simply don't exist. USB? USB-C? They're much, much worse than floppy-disks security-wise actually!

Maybe, by using quantum encryption across a secure channel, but for what purposes and distances?
More...
Not to worry, it will take the USG/US military another 20 years to implement this.:D
 
ironchef said:
I don't think you need to worry about the US, as we still use very "secured" technology:

http://www.npr.org/sections/thetwo-...s-on-outdated-technology-such-as-floppy-disks
More...
Well that must be reassuring for the US's many enemies. I don't suppose they have really bothered too much on that issue.
The only sensible course of action is to annually reduce ALL nuclear arsenals by about 10% per year until they are gone for good.
The current trend of crazy people getting to the top doesn't bode well for planet earth.
 
For day to day chart watching and order entry the man for the job is a computer. Telling the computer exactly how to do that job is difficult though... writing code for that has given me a whole higher appreciation of the human mind. Currently i'm pretty sure that humans are better at developing strategies.
 
You must log in or register to reply here.
Back
Top