AWS re:Invent 2025 - Intelligent security: Protection at scale from development to production-INV214

Scaling Security at Speed: Embedding Expertise, Adapting to Change, and Partnering with the Business

Embedding Security Expertise Throughout the Organization

• Security teams must become builders themselves to secure systems at scale
• AWS has invested in three key areas to embed security expertise:
  1. Primitives: Foundational security building blocks optimized for developer ease of use
  • Example: S2N TLS, an open-source, small, and fast TLS implementation
  2. Embedded Guidance and Tooling: Integrating security across the entire development lifecycle
  • Example: Automated API security testing to eliminate testing toil for developers
  3. Adapting the Security Team Experience: Using internal tooling and AI to scale security operations
  • Example: Automating compliance assessments and active defense systems

Adapting to Changing Risks and Development Practices

• Threat actors are leveraging AI to generate more targeted and convincing attacks
  • Example: Watering hole campaigns using prompt injection and evasion techniques
  • Example: Supply chain attacks targeting open-source package registries
• Security teams must adapt their approaches to keep pace with these evolving threats
  • Measuring the right metrics, like time to fix vs. just number of findings
  • Leveraging AI and automation to increase speed and scale of security responses

Partnering with the Business to Solve Real Problems

• Security teams must deeply understand the business's goals and constraints
  • Example: AWS's focus on scale and rapid development pace
• Embedding security expertise throughout development workflows and operations
  • Example: Automating vulnerability assessments and risk prioritization
  • Example: Threat modeling and securing third-party integrations with business partners
• Using AI and agents to turn security "intentions" into scalable "mechanisms"
  • Example: AWS Security Agent providing contextual security guidance during design and code reviews

Key Takeaways

• Embed security expertise throughout development and operations, not just "shift left"
• Adapt security practices to keep pace with evolving threats and development changes
• Partner closely with the business to solve their real problems, not generic security tasks
• Leverage AI and automation to increase speed, scale, and effectiveness of security

Technical Details and Business Impact

• AWS security teams handle:
  • 312 trillion network flows per day
  • 550 million malicious activities per day
  • 200,000 malicious domains per day
  • 5 billion scans blocked per day
• Lily uses threat modeling and partnerships with AWS to secure their supply chain
• AWS Security Agent can analyze design docs in under a minute, identify security issues, and provide remediation guidance