EFFECTED USERS: Anyone who uses AI, or plans to use it someday.

Hair on fire 4 out of 5

SUMMARY:

In this episode of DIY Cyber Guy, David Schropfer sits down with tech entrepreneur and Deep Sentinel CEO Dave Selinger for a real-world look at how AI is changing the game in security—both physical and cyber. Selinger shares how his company combines smart cameras, deep learning, and live guards to stop crimes before they happen, and why proactive security is more important than ever.

The conversation dives into everything from startup lessons to the future of AI-driven cyber attacks (yes, autonomous cyber weapons by 2027 is on the table). They break down how hackers are already using AI to exploit human error, impersonate loved ones in scams, and push the limits of what is possible online.

Whether you are a cybersecurity pro or just trying to stay ahead of the latest threats, this episode offers eye-opening insights and practical advice for protecting yourself in an AI-powered world.

SHOW NOTES:

In 1901, an American psychologist named Willard Small was the first person to use a rat in a maze for scientific study. Working at Clark University, Small introduced mazes to study learning and problem-solving in rodents, inspired by observations of rats navigating natural environments. He designed a wooden maze modeled after the layout of a Hampton Court garden maze to test how rats learn and remember paths to reach food rewards. His results showed that rats could learn the maze through trial and error, gradually improving their time and efficiency, which helped establish behaviorism and the use of animal models in experimental psychology.

In the test, the rat had to navigate a complex wooden maze with multiple paths, dead ends, and turns to reach a food reward placed at the goal. The rat solved the problem through trial and error—exploring different routes, learning from mistakes, and gradually remembering the correct path over repeated attempts. As the trials progressed, the rat made fewer errors and reached the food more quickly, demonstrating its ability to learn and retain the maze layout.

What if the ‘goal’ was to compromise your computer, and the ‘rat’ was AI? that last paragraph would read:

AI will navigate complex CYBERDEFENSES with multiple paths, dead ends, and turns to reach the GOAL. AI will solve the problem through trial and error—exploring different routes, learning from mistakes, and gradually remembering the correct path over repeated attempts. As the trials progressed, AI will make fewer errors and reached the GOAL more quickly, demonstrating its ability to learn and retain the CYBERDEFENSE layout.

How soon could this be our reality?

Here with me to discuss is Dave Selinger.

Dave is the founder and CEO of Deep Sentinel, a company redefining home and business security with AI and live monitoring. A tech entrepreneur and AI expert, David previously co-founded Redfin and RichRelevance, and he is a recognized leader in machine learning and computer vision.

Welcome.

Is this really on the horizon

How soon

Social engineering.

youtube.com/

westcoast machine learning