A form of Criminal AI is happening now. Botnets have a certain amount of intelligence, Criminal hackers are using sophisticated botnets to perform their functions on the Internet. I want to assert that the precursor to AI (Artificial Intelligence) has been the various programs hackers have developed in the near past.
To truly assess AI we need to define it, and where else to do that then at Stanford’s Computer Science Dept page “What is Artificial Intelligence?”
Definition from Basic Questions page:
It is the science and engineering of making intelligent machines, especially intelligent computer programs. It is related to the similar task of using computers to understand human intelligence, but AI does not have to confine itself to methods that are biologically observable.
With that definition there have already been Criminal AI programs, such as DDOS(Dynamic Denial of Service) programs.
DDOS is essentially ‘intelligent’ just not artificial, as it is run by the criminal. True AI will learn on its own, which would be the next level of Super Virus – Super infections – or worse. In the above diagram the attacker gets the program started and controls what will be the target, but who is to say that an AI program will not learn this methodology to run specific objectives on it’s own? Like attacking all .mil targets? or .cn, .ru etc. it depends on the goals (political , military, economic).
It is only a matter of time before code self replicates and creates it’s own attack vectors. By using it’s initial goals and changing them.
As I mentioned in the past if 10% of the computers are not patched correctly(as in this post) an AI brain will use these to infect and run it’s objectives. Just like the criminals have done this to gain their objectives(usually Ransomware Dollar$).
So what could happen if the criminals move to more sophisticated AI programs rather than the basic ones so far?
You mean like outlined in this Computerworld article?
{Weingarten said it’s possible these rent-a-hacker services may eventually incorporate A.I. technologies that can design entire attack strategies, launch them, and calculate the associated fee. “The human attackers can then enjoy the fruits of that labor,” he said.}
Image from Computerworld article.
So this reality is already happening to a more advanced level. As the Stanford Comp Sci website mentions AI is not just 1 specific intelligence, there can be gradients of intelligence. Such as a 2 year old’s understanding of computers, and on to a teenagers understanding, up to an adult with higher consciousness. The higher levels are not really available and this is where Elon Musk has cautioned we need to be careful where we go in the future of AI in this article of Vanity Fair.
{ Musk believes that it is better to try to get super-A.I. first and distribute the technology to the world than to allow the algorithms to be concealed and concentrated in the hands of tech or government elites—even when the tech elites happen to be his own friends, people such as Google founders Larry Page and Sergey Brin.}
Admittedly there are many schools of thought as to what will happen in AI in the future. The problem is the initial advanced versions of AI … who will control them? a Marcus Aurelius? or Caligula?
The answer to this question will likely determine what will happen with advanced AI in the world.
In the meantime we need to get ready and develop anti-criminal AI developments. Just patching your computer would help. Developing a security departments and figure out what is attacking you are also good steps.
Contact us to discuss advanced steps.