Tarek Cheikh
Founder & AWS Cloud Architect
2025 was a landmark year for AWS security. From SCPs finally getting full IAM language support to post-quantum cryptography going production-ready, here is every major change and what it means for your architecture.
Amazon Inspector expanded from a pure vulnerability scanner into a full application security platform. The new capabilities include:
What to do: If you use Inspector, enable all scan types. You are now getting SAST, SCA, and IaC scanning at no additional cost. Inspector is no longer just a CVE scanner - it is a full AppSec tool.
This is the biggest access control improvement in years. Service Control Policies now support:
You can now write SCPs like: "Deny s3:DeleteBucket only for production buckets" or "Allow ec2:RunInstances only when tagged with CostCenter." Previously impossible.
What to do: Review and upgrade your SCPs. The full IAM language enables surgical precision - rewrite broad guardrails with specific conditions and resource targeting.
AWS Security Hub CSPM now supports the CIS AWS Foundations Benchmark v5.0 with 40 automated controls. This is the latest compliance standard for evaluating your AWS security posture.
The AWS Security Specialty certification exam was updated to SCS-C03 with restructured domains:
The previous version (SCS-C02) expired December 1, 2025. If you are studying for the exam, target SCS-C03.
Extended Threat Detection (originally launched at re:Invent 2024) was expanded to add EC2 and ECS attack sequence findings. Instead of 15 separate findings, GuardDuty now correlates events into a single attack timeline:
credential compromise → lateral movement → privilege escalation → data exfiltration
This is ML-based correlation that detects multi-stage attacks across your compute workloads.
Security Hub now supports up to 1 year of historical trend data with period-over-period analysis, severity filtering, and cross-region aggregation. It has evolved from a finding aggregator into a proper security posture management (CSPM) tool.
AWS implemented NIST-standardized post-quantum algorithms (ML-KEM, ML-DSA) across multiple services:
What to do: Start with CloudFront post-quantum TLS for zero-effort protection against harvest-now-decrypt-later attacks. For long-lived encrypted data, evaluate KMS post-quantum key types.
Graviton5 processors launched with M9g instances: 192 cores per chip and 25% higher performance. Not a security update per se, but relevant for compute-intensive security workloads (log analysis, ML-based detection).
HashiCorp deprecated CDKTF (CDK for Terraform). Existing projects continue to work but receive no updates or security patches.
What to do: If you use CDKTF, start migration planning now. Your options are native Terraform HCL or AWS CDK. No security patches means accumulating risk over time.
CIS released the AWS Foundations Benchmark v6.0 - the latest compliance standard for AWS security posture. Check Security Hub for support status.
For deep dives on specific topics covered here, check out our comparison guides:
This article is just the start. Get the full picture with our free whitepaper - 8 chapters covering IAM, S3, VPC, monitoring, agentic AI security, compliance, and a prioritized action plan with 50+ CLI commands.
Stop sending your IAM policies, CloudTrail logs, and infrastructure code to third-party APIs. Run LLMs locally with Ollama on Apple Silicon — private, offline, fast. Complete setup guide with AWS security use cases.
We obtained the actual compromised litellm packages, set up a disposable EC2 instance with honeypot credentials and mitmproxy, and detonated the malware. Full evidence: fork bomb, credential theft in under 2 seconds, IMDS queries, AWS API calls, and C2 exfiltration.
A deep technical breakdown of how threat actor TeamPCP compromised Trivy, pivoted to LiteLLM, and turned a popular AI proxy into a credential-stealing weapon targeting AWS IMDS, Secrets Manager, and Kubernetes.