Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Saturday October 09 2021, @01:31PM   Printer-friendly
from the The-Adolescence-of-P-1 dept.

Researchers create 'self-aware' algorithm to ward off hacking attempts:

It sounds like a scene from a spy thriller. An attacker gets through the IT defenses of a nuclear power plant and feeds it fake, realistic data, tricking its computer systems and personnel into thinking operations are normal. The attacker then disrupts the function of key plant machinery, causing it to misperform or break down. By the time system operators realize they've been duped, it's too late, with catastrophic results.

The scenario isn't fictional; it happened in 2010, when the Stuxnet virus was used to damage nuclear centrifuges in Iran. And as ransomware and other cyberattacks around the world increase, system operators worry more about these sophisticated "false data injection" strikes. In the wrong hands, the computer models and data analytics – based on artificial intelligence – that ensure smooth operation of today's electric grids, manufacturing facilities, and power plants could be turned against themselves.

Purdue University's Hany Abdel-Khalik has come up with a powerful response: to make the computer models that run these cyberphysical systems both self-aware and self-healing. Using the background noise within these systems' data streams, Abdel-Khalik and his students embed invisible, ever-changing, one-time-use signals that turn passive components into active watchers. Even if an attacker is armed with a perfect duplicate of a system's model, any attempt to introduce falsified data will be immediately detected and rejected by the system itself, requiring no human response.

[Source]: Purdue University


Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 0, Touché) by Anonymous Coward on Saturday October 09 2021, @02:08PM (6 children)

    by Anonymous Coward on Saturday October 09 2021, @02:08PM (#1185768)

    And all your Stuxnet problems will go away.

    • (Score: -1, Troll) by Anonymous Coward on Saturday October 09 2021, @07:59PM (4 children)

      by Anonymous Coward on Saturday October 09 2021, @07:59PM (#1185821)

      Um, why was that modded down? Stuxnet came from Israel

      • (Score: 0) by Anonymous Coward on Sunday October 10 2021, @01:01AM (3 children)

        by Anonymous Coward on Sunday October 10 2021, @01:01AM (#1185864)

        Because it was a JOINT United States and Israel project. Blaming this on "the jews" is disingenuous trolling.

        Get it right, or shut up.

        • (Score: -1, Troll) by Anonymous Coward on Sunday October 10 2021, @01:52AM (2 children)

          by Anonymous Coward on Sunday October 10 2021, @01:52AM (#1185865)

          Blaming this on "the jews" is disingenuous trolling.

          You're full of it. Israel is responsible for many global network attacks... I don't care if it's done "jointly" with other nations. Isreal made Stuxnet. Israel is a Jewish state. We can blame it on "the Jews"

          • (Score: 0) by Anonymous Coward on Monday October 11 2021, @01:20AM (1 child)

            by Anonymous Coward on Monday October 11 2021, @01:20AM (#1186027)
            That's like saying you "made" Facebook, because you're a shareholder/owner via your 401k. True, but utterly obscuring the level of involvement. Israel made half, not the whole thing.
            • (Score: 0) by Anonymous Coward on Tuesday October 12 2021, @07:08AM

              by Anonymous Coward on Tuesday October 12 2021, @07:08AM (#1186366)

              Israel made half, not the whole thing.

              So that makes it okay then? Interesting... Fine, Jews and Protestants.. Happy?

    • (Score: -1, Troll) by Anonymous Coward on Saturday October 09 2021, @09:36PM

      by Anonymous Coward on Saturday October 09 2021, @09:36PM (#1185837)

      Or, stop your stuxnet and all jews will go away.

      if only

  • (Score: 0) by Anonymous Coward on Saturday October 09 2021, @02:26PM

    by Anonymous Coward on Saturday October 09 2021, @02:26PM (#1185774)

    We will all be used like an IoT device, slaves to big brother.

  • (Score: 2, Interesting) by Anonymous Coward on Saturday October 09 2021, @04:35PM (10 children)

    by Anonymous Coward on Saturday October 09 2021, @04:35PM (#1185792)

    The university should be ashamed for allowing such journalistic logorrhea on their server.

    From the few bits of, supposedly, information all-but-drowned by the bullshit river, the guys supposedly add a steganographic signature to the sensor data. Given they felt the need to pad that supposed achievement with pages and pages of pure bullshit, I heavily doubt the thing can ever work.

    • (Score: 2, Insightful) by Anonymous Coward on Saturday October 09 2021, @04:44PM (8 children)

      by Anonymous Coward on Saturday October 09 2021, @04:44PM (#1185793)

      It seems to me that the solution is far simpler than this. Don't connect these freaking things to the internet. I'm not sure why that is so hard to do with such important systems. If it's that important, you can pay somebody to be on site to deal with such things and to go to the site if anything needs to be done.

      • (Score: 0) by Anonymous Coward on Saturday October 09 2021, @04:52PM (1 child)

        by Anonymous Coward on Saturday October 09 2021, @04:52PM (#1185796)

        Is it that hard to use provably secure (Ada, Rust, MITRE C) access points with either OTP or symmetric encrypton keys, plus user authorization, ideally, but not necessarily over direct lines to a centrally managed (and physically manned) securty center? I mean there are lots of straightforward tricks to a secure system and secure handshaking for authorized systems that should keep all but the spook-level groups out, and if they get in you can be pretty sure a human resource or physically tampered device that caused any future security breaches.

        Having said that, we need to get our governments to put security first and exploitation second if we want this to be true.

        • (Score: 2, Informative) by Anonymous Coward on Saturday October 09 2021, @07:15PM

          by Anonymous Coward on Saturday October 09 2021, @07:15PM (#1185816)

          Even if you are using such things, all it takes is an accidental, or intentional, misconfiguration and all that goes out the window. At least with proper airgapping, the damage that can be done is typically much less. Even then, as long as you prevent the viruses from being able to damage physical hardware, you can probably just restore from backup.

      • (Score: 3, Insightful) by maxwell demon on Saturday October 09 2021, @06:19PM (5 children)

        by maxwell demon (1608) on Saturday October 09 2021, @06:19PM (#1185810) Journal

        If you think that not connecting to the internet automatically makes you secure, think again. Hint: Viruses already existed and spread on computers that had not even the hardware to connect to a network.

        Also, how is not connecting to the internet going to secure your systems from insider threats?

        --
        The Tao of math: The numbers you can count are not the real numbers.
        • (Score: 1, Insightful) by Anonymous Coward on Saturday October 09 2021, @07:11PM (2 children)

          by Anonymous Coward on Saturday October 09 2021, @07:11PM (#1185815)

          It doesn't automatically make you safe, but if you don't properly air gap, then there's not much point in securing targets as desirable to hit as the centrifuges being used to enrich uranium. You're also not going to have petabytes of data just disappearing off your systems like had to DARPA a while back if you're having to access and save it to USB devices and smuggle those out.

          Obviously, if you're not screening the things you stick into the system you can still get viruses, but the viruses can't do anything interactive in that environment and you should be able to identify who it is that's in your system screwing things up if you can't launder the traffic through Russia or China.

          • (Score: 5, Informative) by maxwell demon on Saturday October 09 2021, @09:59PM (1 child)

            by maxwell demon (1608) on Saturday October 09 2021, @09:59PM (#1185843) Journal

            As far as I remember, I had read back then that the centrifuges were separated from the internet. That didn't help.

            --
            The Tao of math: The numbers you can count are not the real numbers.
            • (Score: 0) by Anonymous Coward on Sunday October 10 2021, @02:11AM

              by Anonymous Coward on Sunday October 10 2021, @02:11AM (#1185872)

              They were, but then free flash drives!!!11

              Freebies are Teh All-Powerful.

        • (Score: 5, Insightful) by edIII on Sunday October 10 2021, @12:10AM (1 child)

          by edIII (791) on Sunday October 10 2021, @12:10AM (#1185858)

          How did those viruses get introduced into those systems? Outbound modem connections? Data being fed into the system via tape or disks?

          If you don't have an Internet connection, it does automatically make you more secure. Having zero data inputs (tape, disk, USB, bluetooth, wireless) also makes you more secure.

          These systems have to be designed to have a single, highly secure, point of entry into the system. Like the old nuclear silos, you need a team of two people to enter in the same commands into the system for it work. Any outside influences would need to socially engineer two people that are trained to take commands only from a small group of people that they know, or can be authenticated with OTP.

          I wouldn't call OTP signals being layered into the sensor data unadulterated bullshit. Sounds like a pretty effective way to authenticate sensor data, as well physical valves, switches, etc.

          All of this comes down to the willingness to put the levels of security into systems that are a national security risk if they're compromised. That's because executives need bonuses and shareholders need profit, and until it affects them directly, security is unjustifiable expense.

          --
          Technically, lunchtime is at any moment. It's just a wave function.
          • (Score: 2) by Joe Desertrat on Sunday October 10 2021, @08:25PM

            by Joe Desertrat (2454) on Sunday October 10 2021, @08:25PM (#1185971)

            I up modded you particularly for that last sentence. If it is a critical system, the profit motive needs to be eliminated from critical decisions.

    • (Score: 1, Offtopic) by Username on Saturday October 09 2021, @05:31PM

      by Username (4557) on Saturday October 09 2021, @05:31PM (#1185805)

      +5 Informative

  • (Score: 0) by Anonymous Coward on Sunday October 10 2021, @03:07AM

    by Anonymous Coward on Sunday October 10 2021, @03:07AM (#1185885)

    ...NOTHING will go wrong... [youtube.com]

(1)