Skip to content
Scale To Zero by Cloudanix
ST

Guest

Shweta Thapa

Security Specialist Solutions Architect, AWS

Shweta Thapa is a Security Specialist Solutions Architect at AWS, focusing on generative AI security.

Episode with Shweta

What Shweta covered

Designing Security for GenAI With Security Specialist Solutions Architect - Shweta Thapa

Shweta Thapa explains that GenAI security controls must account for non- deterministic outputs, novel attack vectors, and semantic validation. She recommends selecting models from trusted providers, validating both input and output, and keeping context front and centre. Guardrails for input/output, authorisation for data access, and feedback loops for bias adjustment are the practical foundations.

  • While designing Security controls for GenAI applications, a few things should always be kept in mind: 1. Non-deterministic. 2. Different Attack vectors. 3. Semantic validation for output.
  • Some of the basics of GenAI application architecture include 1. selecting a model from a Trusted Provider. 2. Validate Input and Output. 3. Context matters the most.
  • To Secure, utilize Guardrails for input and output. Having authorization in place for data access, and a Feedback loop is key for adjusting bias in output.

New episodes twice a month

No security question left unanswered.

Follow Scale To Zero wherever you listen, or send us the question your team is stuck on and we will put it to an expert.