There is a growing focus on building trust and ensuring responsible AI adoption across enterprises as artificial intelligence continues to transform workplace collaboration. In an exclusive conversation, Sergio Aguilera, Head of Solutions Engineering, APAC, Zoom, speaks with Bhavya Bagga, Business Reporter (Corporate & Leadership), CXO Media & APAC Media, about how Zoom is embedding AI responsibly into its platform, helping enterprises reduce cognitive load, and addressing the key challenges of AI deployment.
One of the biggest concerns with AI adoption is trust. How do we ensure data privacy, accuracy, and responsible use? And how is Zoom addressing these concerns when embedding AI into the platform?
First of all, let me tell you about Zoom’s approach to responsible AI deployments. Zoom is very committed to the deployment and also rollout of responsible AI for our customers considering security, trust, and also privacy and transparency of everything that we do. It’s at the core of everything that we do in terms of developing our products.
Many organizations and enterprises are facing challenges in terms of enrolling and actually rolling out generative AI tools in their organization. So, we take this very seriously at Zoom. We do not use any customer audio, video, chat, screen sharing, attachments, or other communications-like customer content (such as poll results, whiteboard, and reactions) to train our models or any of the third-party models that we actually use in providing our products. In terms of the adoption of AI, we also provide thoughtful controls for our customers.
And what do I mean by that? An administrator can actually manage certain Zoom AI Companion features at the company level, at a group level or a user level having control and granularity over deployment of those features for their account. Furthermore, in terms of transparency, we make it clear when an AI Companion feature is being used in a Zoom meeting. So, we actually provide an in-product notification describing the generative AI Companion capabilities that are activated for that meeting.
What are the biggest challenges that enterprises face when embedding AI into their tasks and how is Zoom addressing those challenges?
I think organizations, as I said earlier, in terms of looking at how to adopt AI tools into the organization, face a number of challenges but I’ll talk about two that are probably very important.
The first is actually defining an enterprise-wide AI policy. If you don’t define an AI policy for an organization, what can happen is a phenomenon called shadow IT, the shadow deployment of AI tools in an organization. Not only is that not consistent for an organization, you end up with duplication, but in particular, you can end up with compliance issues and potentially data leakage. So, it’s very important for an organization to have an enterprise-wide AI policy.
The second is a proliferation of AI tools and vendors like Zoom. There are many tools that customers are having to contemplate looking at. If you don’t have a well-defined AI standard, you will probably struggle with selecting the right AI tools for yourselves.
