Page 44 - Security Today, March/April 2025

E M E R G I N G T E C H N O L O G I E S
manage confi dential data securely and unlock signifi cant value.
This synergy between cloud, AI and cybersecurity offers substan-
tial opportunities for improving systems and data security.
Another factor driving this shift is that many companies have al-
ready been integrating business intelligence (BI) platforms and AI al-
gorithms to analyze their structured data. Now, they are focusing on
unstructured data – video, images, audio and sensor data – to derive
even greater insights, drive growth and enhance business capabilities.
HOW CAN SECURITY STAKEHOLDERS
TAKE ADVANTAGE OF THESE INNOVATIONS
AND HARNESS THEIR DATA IN A PRACTICAL,
ECONOMICAL AND SCALABLE WAY?
Pettersson: Referring back to the three pillars of AI, cloud and cy-
bersecurity, it is important to remember that the camera is the genesis
of any data-driven solution. The camera, or edge sensor, is increas-
ingly important to providing good reliable data for these innovations
to take effect, so the last place to cut corners is at the edge. What is
more, it is not going to be economically feasible or scalable for every-
one to centrally invest in the processing capability. Users will need
distributed intelligence with good devices at the far edge to offl oad
the requirement of that very heavy centralized infrastructure.
You cannot just send everything to the cloud – there is an
economy at scale here. There is a point at which the number of
cameras or devices becomes too large for the cloud to effectively
manage. While the cloud is useful for experimentation and scal-
ing, a hybrid environment becomes crucial beyond this threshold.
Of course, there will be a need for small systems to go right from
the device to the cloud, and some very secure environments will
have to do everything on prem, but we will see the vast majority
of our market go to a hybrid environment depending on scale,
competence and need for privacy and data control.
Trinh: For systems integrators (SIs) taking a practical ap-
proach to this journey, it is crucial to leverage vendor relationship.
Ask vendors to provide quantifi able metrics that demonstrate the
value of specifi c technologies to end customers. For example, for
AI models that impact computer vision, what are the minimum
detection and classifi cation thresholds the vendor’s analytics pro-
vide? Additionally, investigate potential integrations into other
systems that are not physical security-related, such as enterprise
resource planning (ERP), fi nance & operations and BI systems.
WHAT TOOLS HARNESS NEW INNOVATIONS
FOR EVERYONE IN THIS VALUE CHAIN FROM
THE DEVELOPER TO THE SI TO THE END USER?
Pettersson: From a device manufacturer’s standpoint, we need to
easily integrate into the commonly used development environ-
ment where developers work. This requires a developer suite with
open integrations into third-party systems and platforms including
those from mainstream tech players like Google Cloud, AWS and
Azure. Key factors for success include open architecture for seam-
less integration, scalability to meet growing demand, and a strong
Malambo C/peopleimages.com/stock.adobe.com
4 4 M A R C H / A P R I L 2 0 2 5 | S E C U R I T Y T O D A Y