How we might be able to ruin art, save society and speak to whales – Source: www.cybertalk.org

Robert Falzon is currently the Head of Engineering, Canada, within the office of the CTO for Check Point Software Technologies Inc., the world wide leader in securing the internet. His background includes over 30 years of experience in large scale network security architecture, design, and deployment projects for government and business organizations spanning the globe. Currently leading a large team of the most talented cyber security engineers in the industry, Robert and his team are responsible for educating the market on the latest cyber security trends.

Artificial intelligence has the potential to enhance humanity’s beauty, and yet the uncertainty around its contours casts a shadow over its value. In this interview, Head of Security Engineering, Rob Falzon, puts AI’s potential into perspective, particularly in relation to art, the emotional life, and cyber security.

These thought provoking insights will shift how you see AI and what it stands to offer. Starting with the rose-colored glasses and moving towards the futuristic and Frankenstein, AI will never appear the same to you again.

How do you see AI as “opening the aperture” in terms of humanity’s capabilities?

Good question. As we see it today, AI already frees us up to do many things that we would otherwise needlessly sink time into. It automates routine tasks for us. It helps to answer simple questions. It makes us more accurate because it can provide us with detailed information very rapidly. In turn, this enables us to expedite decisions that would typically take us a lot longer to make. Having these AI inputs can be very helpful.

AI also helps us automate some creative tasks, and it enables us to be more innovative, I think, as the heavy-lifting is sort of being taken care of already. That’s the early stage of what we’re seeing AI do for us – it’s helping to free us from select manual tasks and allowing us to be more creative and innovative.

As a photographer, how do you see artificial intelligence impacting the creation and production of art?

AI has had an immense impact in a very immediate way. As soon as we started seeing things like deepfakes and similar types of content hit the art scene, people were pretty shocked. But it was interesting from an artistic perspective.

Of course, there’s art that your kids come home with, that you put on the fridge, right? And about which you say, ‘Wow, it’s amazing’. But then you look at something like advanced photography, and you go, ‘Well, what is it that makes that photograph compelling?’

And oftentimes, the ‘what makes it compelling’ is not the photo itself. Rather, it’s the ability to feel that photo and to say, ‘Wow, I can discern the feeling that’s being conveyed in that image; there are emotions invoked by looking at it.’

Early on, what we saw was that machines were having a very difficult time replicating that human emotional connection. But now, that’s actually changing.

Despite the fact that there are still technological weaknesses — in relation to art, a lot of artificial intelligence has problems with things like rendering human hands and feet — it’s getting much, much better.

In the art world, AI also struggled to generate emotional connections with people because we need for people to look real – sometimes you’re not looking for a person with perfect skin or perfect symmetry in their face. Machines would generate a person that looked perfectly symmetrical, and that takes away from the realism of the art, as that actually doesn’t come across as a real human being.

Now, I’m starting to see the latest releases of some of the software I’ve been working with; Stable Diffusion is one SDXL that I’ve been looking at for the last little while. And it’s now creating realistic images of people that are very much believable — it is connecting more on an emotional level.

For example, I’ve seen AI-generated pictures of animals sitting in the rain, looking sad. And when I see those images, I feel like ‘wow’ it makes me really sad to see that animal.’ So, recent developments in AI are really having a genuine impact on humans, in many ways.

Artificial intelligence is also creating a challenge, I think, when it comes to artistic attribution. Because, you know, if anything can be created with anyone’s style, it makes it difficult for the original artist to obtain credit for the work that he/she has done.

Remember, all of these things are actually stealing tidbits of that art from the internet — anything that’s been posted as collective. That’s the ‘feeding material’ if you will, for these systems. So, AI is having an immense impact, I think, on art in general.

Do you think that AI will lead to the devaluation of real artists and their works, as AI will be able to render images more clearly, precisely and beautifully than humans?

I can already see the issue in terms of the devaluation of artists. That’s something that I alluded to previously when I mentioned that attribution is becoming more difficult.

An artist might spend years trying to develop a style that’s their own. If you think about famous photographers, like Ansel Adams, and folks like that, they have a unique style. Now, if I can just tell the machine to create an image for me in that style, that does take something away from the result.

I think it makes it far more difficult for us to benefit from the creation of that art and it sort of begins to remove the option of art as a way for someone to earn an income. So that does create a challenge. I personally have seen that with friends of mine who are professional photographers, who are now telling me that the agencies that they used to work with are having trouble getting work. Organizations find it less expensive to just tell an AI to create something for them. And then of course, organizations have the rights to the AI-generated photos anyway, meaning that they don’t have to pay for the work.

AI is having an impact where it is devaluating artists’ capabilities – for sure.

In our pre-interview, you mentioned that AI preys on our weakest points (our humanity). Can you please speak to that a bit further?

We talked about AI as ‘doing something’. And I think that’s the challenge — to think about ‘well, is AI really doing that or is there a bad actor behind that activity?’ Oftentimes, it’s the latter.

AI and large language models are tools that are being used for good, like creating art, but they haven’t quite reached a level where they can determine what they should be doing and not doing for themselves.

And they have not yet developed morals, or anything like that. They’re only reflecting what their training has provided them with. So, if I think about how AI preys on human weaknesses, imagine:

Right now, elderly people are sometimes easy targets in phone scams. Someone will call and say ‘hey, you know, your machine’s not working properly. Help me connect to your computer and I’ll help you with it’. And that someone is very convincing, and the elderly person gets them onto their device. And eventually, the threat actor ends up scamming the victim and taking money from their bank account, or what have you.

These scams have already been very successful. But shift your frame of mind, and now imagine that the same scam is super-powered by artificial intelligence. Instead of grandma getting a phone call from a random stranger who’s talking about their computer, it might be a call from their daughter, son, close relative or loved one, saying ‘hey, this isn’t working correctly.’ Or ‘hey, I really need your help. I’m stuck in [fill-in-the-blank] and if I don’t get $1,000, they’re going to put me in prison.’

And in reading this, you might think that it’s just a ransom person saying those things — but it’s not, as it’s actually using the voice of their loved one. So, artificial intelligence is being used in machine learning; beyond art, it’s being used to imitate people, their voices, their behaviors and even the nuances of how they speak.

I recently saw a long presentation by former U.S. President Barak Obama that was entirely generated by artificial intelligence. And it’s shocking as to how realistic the presentation was.

AI is super-powering attacks. I think that’s the scary part in terms of preying on our weakest points; AI is now preying on our emotions. Humans are susceptible, as humans are at their weakest when pushed into seemingly urgent circumstances — when emotions are at their peak.

Previously, you’ve mentioned that AI stands to become a modern “Galileo’s telescope. Would you like to speak to that a bit further, please?

Yeah, that’s one of my favorite things. Think about Galileo and how he had this telescope and he sort of peered into a completely unknown universe — but he had an idea of what was going on, yet he didn’t really know for sure. He needed some sort of proof. And by developing that telescope and increasing the magnifying power, he allowed the rest of us to see into the world around us, which was completely opaque to us before.

Artificial intelligence really does have the same opportunity to create that for us — I was reading about how artificial intelligence is now being used to develop the ability to speak to animals. And you might be reminded of the stories that we used to read as little children about speaking to the animals, and Dr. Doolittle, and things like that. But the ability for AI systems to be able to identify minute patterns in vast amounts of data has provided us with an opportunity to literally potentially begin speaking to whales in order to find out about what’s harming them or how what we’re doing is affecting them, or what they know, and so forth.

Now, there are all sorts of ethical concerns about doing that. I’d be very concerned about how people approach that. But there’s really an opening for humanity here — there’s another world that we know is there, but can’t currently see into. AI may change that.

Nevermind AI’s ability to, as I said earlier on, free us up to do some of the more complex thinking and innovative stuff — the medical side of it is incredible. These tools can identify cancers long before a trained professional can.

We’re now seeing an opportunity the people who are brilliant among us — think of people like Steven Hawking — to have more fulfilled lives and to have more opportunities to share their gifts with us. I do think that these tools are opening up opportunities to see a world around us that was not directly available to us before.

As we’ve touched on previously, AI has taken the world by storm, in-part because it offers a sense of optimism (much needed on the heels of a pandemic). What benefits of AI capture your imagination, both personally and professionally and why?

I believe that we’ve reached a precipice. All of this is developing so quickly. I’ve been in technology all of my life. My career at Check Point spans 20 years and before that, I was all over Europe and the Middle East, but always associated with technology.

To me, in seeing this technology advance over that time frame, you sort of get an idea of how things move in technology. The last five years has completely blown things apart.

It seems like every day, I wake up and I read something new that’s amazing. It’s absolutely amazing. Sometimes, I’m saddened by the fact that I’m probably not going to live long enough to see the revision of this. But as I think about the opportunities that my son is going to have, and the types of roles that are going to be available to people like him in the future, I’m excited by it.

I’m also excited by the fact that artificial intelligence has an immense opportunity to create equity for humanity. There’s the potential for AI to be used to reach remote areas or underprivileged communities. Again, I think about the healthcare aspect of it, and being able to provide better healthcare to those who may not be able to afford the high premiums that most providers are charging nowadays.

I think about the ability to easily connect with people who speak other languages. I think that AI could go a long way towards creating global peace, provided that it enables us to find means of obtaining better, cheaper energy. I think AI has a lot of potential. And I hope that even when I’m gone, that potential continues to grow.

What are your thoughts on AI and cyber attack techniques?

It has ‘supercharged’ attacks. The same kinds of attacks are happening as before, but AI has enabled them to become more advanced and more menacing.

At Check Point, we are developing and we will continue to develop advanced tools that help protect people from advanced cyber security challenges.

For more AI insights, please see CyberTalk.org’s past coverage. Lastly, to receive more timely cyber security news, insights and cutting-edge analyses, please sign up for the cybertalk.org newsletter.