In this episode of The BlueHat Podcast, host Nic Fillingham and Wendy Zenone share David Weston’s keynote from BlueHat India 2025. David explores the growing role of on-device AI in Windows, the security risks it introduces, and how Microsoft is rethinking architecture to defend against new threats like model tampering, data exfiltration, and AI-powered malware. He also shares insights on innovations like Windows Recall, biometric protection, and the future of secure, agentic operating systems. 

In This Episode You Will Learn:  

• 
  

  How AI integration in Windows (like Windows Recall and MS Paint) is evolving 

  
  

• 
  

  Emerging threats from protocols like MCP and CUAs 

  
  

• 
  

  What a “confused deputy” attack is, and how Microsoft is protecting users 

  
  

Some Questions We Ask: 

• 
  

  What are the biggest security threats in on-device AI—data, model, or runtime? 

  
  

• 
  

  Can AI be used to accelerate post-compromise attacks? 

  
  

• 
  

  What will it take to bring Azure-level confidential computing to the consumer device? 

  
  

Resources:      

View David Weston on LinkedIn   

View Wendy Zenone on LinkedIn   

View Nic Fillingham on LinkedIn  

Related Microsoft Podcasts:   

• 
  

  Microsoft Threat Intelligence Podcast   

  
  

• 
  

  Afternoon Cyber Tea with Ann Johnson   

  
  

• 
  

  Uncovering Hidden Risks   

  
  

Discover and follow other Microsoft podcasts at microsoft.com/podcasts   

The BlueHat Podcast is produced by Microsoft and distributed as part of N2K media network.