Chapter 14: Building Secure and Reliable Systems
Is the code running in your production environment the code you assume it is? Your system needs controls to prevent or detect unsafe deployments: the deployment itself introduces changes to your system, and any of those changes might become a reliability or security issue. To keep from deploying unsafe code, you need to implement controls early in the software development lifecycle. This chapter begins by defining a software supply chain threat model and sharing some best practices to protect against those threats. We then deep dive into advanced mitigation strategies such as verifiable builds and provenance-based deployment policies, and conclude with some practical advice about how to deploy such changes. Previous chapters addressed how to consider security and reliability when writing and testing your code. However, that code has no real impact until it’s built and deployed. Therefore, it’s important to carefully consider security and reliability for all elements of the build and de
Chapter 14: Building Secure and Reliable Systems Chapter 14 Deploying Code By Jeremiah Spradlin and Mark Lodato with Sergey Simakov and Roxana Loza Is the code running in your production environment the code you assume it is? Your system needs controls to prevent or detect unsafe deployments: the deployment itself introduces changes to your system, and any of those changes might become a reliability or security issue. To keep from deploying unsafe code, you need to implement controls early in the software development lifecycle. This chapter begins by defining a software supply chain threat mod
Explore this link on the map →related reading
- When AI Writes the World's Software, Who Verifies It? — Leonardo de Mouraleodemoura.github.io
- Your job is to deliver code you have proven to worksimonwillison.net
- Defining A Software Supply Chain Security Platform & Exploring New Techniques, Part 2softwareanalyst.substack.com
- Auto-review of agent actions without synchronous human oversightalignment.openai.com
- A shallow dive into formal verificationvitalik.eth.limo
- 2312.06942arxiv.org
- Turnkey Whitepaperwhitepaper.turnkey.com
- Software Supply Chain Security (Part 1)softwareanalyst.substack.com
- AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
- Treat Agent Output Like Compiler Output | Skipskiplabs.io
- How we contain Claude across products \ Anthropicanthropic.com
- Responsible Scaling Policy Updates \ Anthropicanthropic.com