Episode
Episode 51: From London to Silicon Valley: AI Security for the Enterprise with Harmonic's Alastair Paterson
- Podcast
- 909Exec
- Published
- Dec 29, 2025
- Duration seconds
- 2625
- Processing state
not_requested
Actions
POST https://stenobird.com/v1/public/podcasts/909exec-7089471/episodes/episode-51-from-london-to-silicon-valley-ai-security-for-the-enterprise-with-harmonic-s-alastair-paterson/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/909exec-7089471/episode-51-from-london-to-silicon-valley-ai-security-for-the-enterprise-with-harmonic-s-alastair-paterson.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
DESCRIPTION: Join Den Jones as he sits down with fellow Brit Alastair Paterson, CEO and co-founder of Harmonic Security, to explore the challenges of AI adoption in enterprise environments. From bootstrapping a cybersecurity startup in London to scaling in Silicon Valley, Alastair shares insights on securing AI workflows, the cultural differences between UK and US startup ecosystems, and why traditional DLP solutions fall short in the AI era. GUEST INFORMATION: Name: Alastair Paterson Title/Credentials: CEO and Co-founder of Harmonic Security Background: Former CEO and co-founder of Digital Shadows (acquired in 2022), cybersecurity entrepreneur who moved from London to San Francisco to scale his companies [3:03] Connect with guest: Based in San Francisco, actively hiring across 20 open positions [31:25] EPISODE HIGHLIGHTS: Introduction to Harmonic Security Harmonic enables secure AI adoption for enterprises, sitting between business teams wanting to leverage AI and security teams managing risk [1:43] Key quote: "Huge amount of tension between the business that's wanting to push forward and lean into this wave in a big way. And then the security legal and compliance teams are trying to hold the tie back a little bit" [1:59] The Evolution of Harmonic's Mission Originally focused on data protection using language models, evolved to address broader AI governance challenges [5:32] Key quote: "We can start to use language models for data protection too... with the context and understanding around what users are doing and the intent behind the data" [5:17] Why Traditional DLP Fails Traditional DLP solutions are inadequate for modern data protection needs [7:00] Key quote: "DLP has always sucked. It's been really just a money pit of uselessness" [7:00] Harmonic's Deployment Ap…