A talk by Shashank Golla at Wiz
Notes
- Known threats:
- Front-Door Access
- Excessive IAM permissions and Lateral Movement
- Supply Chain Attack
- Elastic Kubernetes Service exposed, and then exploited in less than 20 minutes
- Deepseek Dataleak
- Over 85% of organizations are using some form of AI: Langchain Hugging face OpenAI Tensorflow
- Threats from vibe coding
- AI as infrastructure, acceleration without control (No AI Governance)
- ASPM systems to check for SAST and DAST
- CSPM Scanner
- How do we know the developer who pushed the code to te cloud? This will increase time of remediation
- For run-time, we use EDR
- Wiz will run risk assessments for everything like:
- Open secrets
- Vulnerabilities
- Identity
- Configuration
- Data and AI
- Public exposure
- Customers will attempt to use Wiz for:
- Proactive approach to run risks assessments of AI Agent
- Runtime security with Wiz Sensor
- Cloud-to-code graph tracing a resource to a repo
- Checking IAM
- Developing security guardrails early in te lifecycle
- Starting from the Integrated Development Environment that Wiz Code acts as a extension, has guard rails
- Wiz security cloud can view from Terraform file, so that you can actually see any exposed cloud services
- Wiz allows for democratization by bringing many different teams to interface with their platform, Wiz is cohesive and will give custom dashboards, they can self-serve the portal themselves.
- Wiz alert, answers ‘so what’ question, tell you why you should fix this, the impact of the vulnerability
- Wiz allows creation of projects to tailor make differnt interfaces
- Services are checked and found for services that are related, We can attach clear ownership of who the developer is, and send alerts directly
- Wiz Lens
- Wiz MCP
- Wiz Defend
- Root cause analysis can find the code owners, using VCS reading
- Click-Ops
- Palo Alto OAuth Token Breach