Latest NVidia GPUs used to accelerate cracking WiFi !


Recommended Posts

When NVIDIA claimed that CUDA on combo with a fast GPU is very powerfull to use for computing, I'm pretty sure they didn't have this in mind. An interesting article over at SC Magazine. Seems a Russian firm is using the latest NVidia GPU's to increase WiFi password recovery by 10000%. Amazing in how powerful GPU's have become in the last few years eh ?

Global Secure Systems has said that a Russian's firm's use of the latest NVidia graphics cards to accelerate WiFi ?password recovery' times by up to an astonishing 10,000 per cent proves that WiFi's WPA and WPA2 encryption systems are no longer enough to protect wireless data.

David Hobson, managing director of GSS, claimed that companies can no longer view standards-based WiFi transmission as sufficiently secure against eavesdropping to be used with impunity. He also said that the use of VPNs is arguably now mandatory for companies wanting to comply with the Data Protection Act.

He said: ?This breakthrough in brute force decryption of WiFi signals by Elcomsoft confirms our observations that firms can no longer rely on standards-based security to protect their data. As a result, we now advise clients using WiFi in their offices to move on up to a VPN encryption system as well.

?Brute force decryption of the WPA and WPA2 systems using parallel processing has been on the theoretical possibilities horizon for some time - and presumably employed by relevant government agencies in extreme situations - but the use of the latest NVidia cards to speedup decryption on a standard PC is extremely worrying.

?The $64,000 question, of course, is what happens when hackers secure a pecuniary advantage by gaining access to company data flowing across a WPA or WPA2-encrypted wireless connection. Will the Information Commissioner take action against the company concerned for an effective breach of the Data Protection Act.

Thanks Zinjin for reporting his news.

source

OMG !

hmm our neighbors got WiFi ,i could use some help from my:devil: :devil:

10000% acc can do wander

10,000% = 100x faster than CPU ... 100% faster would be 2x CPU

Using the methods described in the article, theoretical performance is boosted up to 100x than typical.

The length of time to crack a simple password used for WPA/WPA2 is now going to be vulnerable. Strong passwords and especially, lengthy strong passwords, won't be vulnerable anytime soon.

10,000% = 100x faster than CPU ... 100% faster would be 2x CPU

Using the methods described in the article, theoretical performance is boosted up to 100x than typical.

The length of time to crack a simple password used for WPA/WPA2 is now going to be vulnerable. Strong passwords and especially, lengthy strong passwords, won't be vulnerable anytime soon.

wait so the crack would presumably be still on the passkey itself instead of the algorithm yea? in which case 63 character printable-ascii keys would still be relatively secure.

10k time faster

using gpu

imgine using sli

or heck gpus farm !

Yes, 10k times faster.

But 10k times faster than what?

Just gets me how they can't actually put a time scale on it (yeah, I know, processor speeds vary etc)

I mean my Q6600 is probably at least 10k times faster than a 286, but neither's going to be able to model every particle of matter in the universe in real time.

Yes, 10k times faster.

But 10k times faster than what?

Just gets me how they can't actually put a time scale on it (yeah, I know, processor speeds vary etc)

I mean my Q6600 is probably at least 10k times faster than a 286, but neither's going to be able to model every particle of matter in the universe in real time.

faster then cpu

but you got your point (Y) .... faster then athlon x64 or core 2 duo or top of the line QX9700 that remain untold too bad

The First Time I have Ever Heard about cracking WEP , WPA WPA2 was in 2005. cracking wep only takes about 2-3mins. wpa/wp2 depended on the word list that can take up to 4 to 15gb, would take about a week or two running a top of the line computer 24/7. using a GPU to speed up the process now no one is safe. get your PS3 to Crack wpa/wpa2 in no time if you dont have a top of the line Video card.

The First Time I have Ever Heard about cracking WEP , WPA WPA2 was in 2005. cracking wep only takes about 2-3mins. wpa/wp2 depended on the word list that can take up to 4 to 15gb, would take about a week or two running a top of the line computer 24/7. using a GPU to speed up the process now no one is safe. get your PS3 to Crack wpa/wpa2 in no time if you dont have a top of the line Video card.

The biggest problem doesn't lie with very powerfull machines like PS3/cell servers or clusters for that matter. The main issue is that the average consumer notebook with dedicated graphics cards can now do this. On the go. Hide out in a car outside a corporate enterprise and you're in like that.

That's why I suggest IPSec: http://en.wikipedia.org/wiki/IPsec

It secures the TCP/IP layer and when used with technology such as NAP, it's getting harder.

faster then cpu

but you got your point (Y) .... faster then athlon x64 or core 2 duo or top of the line QX9700 that remain untold too bad

I don't think you get it. The GPU is 10,000 times faster than the CPU, but how long does it take the CPU? They've accelerated cracking, but does it bring it down to a feasible amount of time?

I don't think you get it. The GPU is 10,000 times faster than the CPU, but how long does it take the CPU? They've accelerated cracking, but does it bring it down to a feasible amount of time?

idk

Did anyone really think WPA and WPA2 were so powerfully encrypted? They were bound to be cracked eventually, there is no such thing as an uncalculable number...

I'm just suprised a GPU was able to do this rather than a normal CPU...

gpu are more powerfull then cpu

in decoding & encryption thingies but it luck mostly apps to take advantage of it

usually some bruteforce methods can take a year to complete (around 350 days) now 100x faster your looking at 3.5 days

the theoritically impossible (noone is going to bother running their pc for 350 days to crack a password in that time the password may change :p or something)

has now become possible because not much happens in 3.5 days :)

thats just a rough estimate and to give you a rough idea on the performance gain

from what i was told if your using a really secure password (i.e. something like 63random characters like generated from grc.com , https://www.grc.com/passwords.htm ) you still dont have much to worry about ;)

bottom line = use a password generated from that site ( https://www.grc.com/passwords.htm ) and you wont be hacked ;)

Edited by ThaCrip
This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • 47% profit margin? Wtf!! I know companies are in business to make money but come on man. I know for a fact I'll never own one of these.
    • Most AI-powered mainframe migration vendors expected to fail by 2030, Gartner warns by Paul Hill Credit: Pexels You may have read that many companies still run code written in ancient programming languages like COBOL and pay a handsome sum for those who can maintain that code. Well, it looks like this area of the tech world could be the scene of an AI bubble. It turns out that there are mainframe exit vendors, helping companies move their legacy mainframe systems to modern cloud environments or servers such as Microsoft Azure and AWS, using generative AI tooling. Unfortunately, 75% of these vendors are now expected to pivot or cease operations as market realities take hold by 2030. Alessandro Galimberti from Gartner said: Some of the companies in the mainframe exit market are IBM, 21CS, BMC, Broadcom, Rocket Software, DXC, GTSG, and Kyndryl. The reasons some of these firms are expected to quit the market are a reset of market expectations and a decline in demand for one-size-fits-all migration solutions. The reset in expectations is likely to be driven by cost overruns and threats to business, and the potential occurrence of critical failures within businesses as a result of bad transition implementations. These insights from Gartner are pretty interesting because it’s a specific area of the market where doubt is being cast on generative AI. Many people have cast doubt on whether AI companies will successfully justify the massive amounts spent on GenAI to date, and this data from Gartner suggests the road could be rocky for GenAI.
    • Heaven forbid they lose pennies from their Trillions! Like always, the consumer pays the most. Why is Tim Cooks even talking.....shouldn't he be packing up his office??
    • If you have the budget...! Some solo or indies just want to either learn or start their game and aren't in a capacity to pay salaries or to contractors... Get real.
    • Source and more 35 years old?! And if my maths is mathing, that means she was around 10 when The Ring came out?! Damn...scariest 10 year old I think I've ever seen. 
  • Recent Achievements

    • Week One Done
      Classifyskilleducation earned a badge
      Week One Done
    • One Month Later
      eurospharma62 earned a badge
      One Month Later
    • Week One Done
      With What earned a badge
      Week One Done
    • Week One Done
      Harris Gilbert earned a badge
      Week One Done
    • One Month Later
      Vincian earned a badge
      One Month Later
  • Popular Contributors

    1. 1
      +primortal
      537
    2. 2
      +Edouard
      168
    3. 3
      PsYcHoKiLLa
      76
    4. 4
      neufuse
      64
    5. 5
      ATLien_0
      63
  • Tell a friend

    Love Neowin? Tell a friend!