Vikesh Khanna, Chief Technology Officer and Co-Founder at Ambient.ai reveals what makes the company stand out in the market.
My name is Vikesh Khanna and I’m the Chief Technology Officer (CTO) and Co-Founder of Ambient.ai.
At Ambient.ai, I lead our product and engineering teams – I am responsible for building a world-class product with cutting-edge technology that supports the company’s mission of preventing every security incident possible.
Before founding the business, I built enterprise software products and applied AI systems at Microsoft and Dropbox.
My passion for AI began at Stanford more than a decade ago – I was a graduate student with a research focus in deep learning and computer vision.
We built and witnessed, firsthand, the very first AI systems that surpassed human performance in 2012.
It was a transformative experience.
Since then, I’ve had this deep-rooted conviction that AI will advance rapidly over the coming years and will be indispensable in solving many meaningful problems.
I’ve had the great fortune of working with and speaking to hundreds of wonderful security professionals over the last eight years.
Without exception, every one of them wants to be proactive; they want to prevent incidents – but how do you do that with an incredibly small team monitoring hundreds of cameras and dealing with thousands of false alarms?
Physical security teams have this incredibly important mandate of keeping people, places and assets safe – however, the tools and technology at their disposal forces them into a reactive posture.
I think you would agree that there is no future in which physical security becomes proactive by staffing a room with hundreds of people to watch cameras all day.
Which is why I believe AI not only benefits physical security, it is actually essential to achieving the mandate of being proactive.
AI can monitor all your camera feeds 24/7 for suspicious activity and threats, and direct your operators and guards to what matters.
The technology is now fully capable of separating regular behavior from suspicious activity.
This kind of real-time situational awareness enables your team to run a proactive and autonomous security operation.
Our AI platform is unique in two aspects: the depth of our contextual scene understanding and the breadth of our detection capabilities.
Let me start with context. From the very beginning, Ambient.ai has been built on the conviction that scene context is not just important but critical for threat detection in videos.
For instance, is a person carrying a knife suspicious? It may not be suspicious in a kitchen, but it is most certainly suspicious in a lobby.
Cleaning crews and patrol guards at night are not suspicious, but a masked individual with a crowbar next to your car is most certainly suspicious.
This is why we pioneered the idea of contextual scene understanding, which has been deeply embedded in our product from day one.
Secondly, and equally important, is the breadth of our detection capabilities.
We have the largest repository of threat signatures of any product in the industry – over 150 signatures today and rapidly growing.
Why is this important? Because security incidents are not isolated behaviors – they almost always involve a trail of events that are all opportunities for proactive intervention.
For instance, an individual casing the building, jumping a fence, forcing a door open, walking in through a propped open door, wearing a mask and so on – these are all suspicious behaviors that may culminate in a more serious threat, and they are all signatures in our product.
True security automation goes far beyond just detecting weapons and slip and fall.
So, two things differentiate us from the rest of the market: deep contextual scene understanding and the breadth of our detection capabilities.
This is an incredibly exciting time to be at the intersection of AI and physical security.
More advancement has happened in AI in the last year than in the previous ten years combined.
We’re truly on an exponential curve of AI advancement right now. So, in 2025 we will continue expanding the intelligence capabilities of our platform.
We are investing heavily in AI research and development, particularly in areas like agentic AI and even deeper contextual scene understanding.
We are also actively expanding our offerings to new verticals and geographies.
We will continue to focus on building strong partnerships within the industry, especially with organizations like Security Journal Americas and the Security Industry Association (SIA).
We’re seeing unprecedented interest in AI adoption in the physical security industry right now – I am excited for and looking forward to significant growth and innovation in the coming year.
This article was originally published in the April edition of Security Journal Americas. To read your FREE digital edition, click here.