Secure AI: Build Agents with Confidential Data Access
Nothing is loaded from YouTube until you press play, and then without tracking cookies.
About this video
Discover how to build secure AI agents that can access your data, including Slack, email, and chats, without compromising privacy. We're stacking layers of capabilities on secure hardware, enabling confidential data handling, monetization, and agent marketplaces. It's time to trust AI. #SecureAI #AIagents #DataPrivacy #MachineLearning #ConfidentialComputing #AIsecurity #ArtificialIntelligence #TechInnovation #FutureofAI #AIdevelopment #NEARprotocol
Transcript
143 wordsCaptions as YouTube publishes them, usually machine-generated, so product and people names can be misheard.
- 0:00
Now you can run this model on a benchmark and actually produce results that are uh correct. Then you can build agents on top of it using this model where agents as well are you know using confidential data. You can trust them with actually accessing your slack and your email and your chats because right now you know like giving access to all of your life to you know one of these models is pretty scary.
- 0:24
So we're kind of stacking this layers of uh capabilities on top of this fundamental kind of uh secure enclave hardware. uh allowing to monetize models both open source, closed source, whatever this is, allowing to have confidentiality, allowing to have shared private memory, allowing to have agents and then on top of it you can have agent marketplaces and you know personal assistants and more,