Cloud Security Compare

Buyer note

What changed for CNAPP buyers in 2026: nine months of vendor news

Cloud Security Compare editors · · Editorial assessment

The short version

Compare current capabilities rather than corporate announcements: Wiz documents code scanning and runtime blocking, Cortex Cloud documents application and runtime security, Upwind documents application and AI security, and Defender for Cloud separates posture and workload plans. Confirm availability and packaging in an evaluation.

Which Wiz capabilities should you validate

Wiz Code documents SAST and AI-SAST, SCA/SBOM, IaC, secrets, malware scanning and code-to-cloud remediation. Wiz Defend combines agentless telemetry with the Wiz Sensor for detection, investigation and threat blocking. Validate supported deployments and prevention policies.

For each cloud you run, request the supported workload and service matrix and confirm which code and runtime capabilities the proposal includes. More on the options: Wiz alternatives.

What is Palo Alto Networks adding around Cortex Cloud

Cortex Cloud documents application, posture and runtime security, including attack-path context for vulnerability prioritization. Ask which capabilities and deployment requirements apply to the proposed licence. More: Prisma Cloud alternatives.

Which Upwind capabilities should buyers evaluate

Upwind combines agentless scanning with runtime baselines, forensic timelines and response workflows. Its application-security page describes scanning human- and AI-written code, and its AI-security page describes agent, tool and MCP scanning and runtime monitoring. Our current scores are runtime (4.6), code (4.4) and ecosystem (4.0); validate language coverage, availability and integrations. More: Orca vs Upwind and Upwind vs Wiz.

Which AI features did vendors ship

AI was the most common subject of 2026 announcements, in two different senses: securing AI workloads, and using AI agents to do security work.

  • Orca Security announced AI AppGen Security and AI Code Security Auditor on 29 July 2026, ahead of Black Hat USA. Orca expects general availability later in 2026.
  • Sysdig introduced Headless Cloud Security on 6 May 2026, which delivers CNAPP functions as agent skills inside AI coding agents, and Sysdig Secure AI on 4 August 2026.
  • Aqua Security announced Aqua Compass on 22 April 2026, an MCP server that lets AI agents investigate and contain runtime incidents with human oversight.
  • Microsoft added threat protection for AI agents (preview, 2 February 2026) and AI model security for Azure Machine Learning (preview, 30 March 2026) to Defender for Cloud.
  • CrowdStrike added a Cloud Risk Engine that maps cloud risks to adversary tradecraft, announced on 24 March 2026.

The buyer's question is the same for each: is it generally available, which plan includes it, and can you test it in your proof of concept? Several of these launched as previews or with availability later in the year.

What changed in pricing and packaging

The one dated pricing change came from Microsoft. From 27 October 2026, Foundational CSPM, the free posture tier of Defender for Cloud, is no longer turned on by default for new Azure subscriptions. It stays free, existing subscriptions are not affected, and AWS and GCP connectors are unchanged. Teams that rely on the free tier for new subscriptions need to turn it on themselves. No other vendor in our matrix published a price list this year, so Defender for Cloud remains the only platform here with public pricing, per resource or per Defender Workload Unit. More: Orca vs Microsoft Defender for Cloud.

What did the open-source projects ship

Falco, the CNCF graduated runtime detection project that Sysdig created, released 0.43.0 in January, 0.44.0 in May and 0.45.0 on 21 September 2026. Version 0.45.0 adds raw byte matching in rule conditions and new rules that flag containers accessing GPU devices. Sysdig's runtime score rests partly on Falco, and teams that run Falco themselves should ask which version a vendor's managed rules follow.

Do these changes move our scores

All nine vendors and 63 criterion scores were re-reviewed on 4 November 2026. Current documented capabilities changed some scores, while the published weights stayed fixed. Announcements alone do not establish support or plan availability. The news page lists each announcement with a link to the original source.

Sources

Related pages