.svg)
In the space of one week, OutSystems shipped two governance-focused platform updates, custom AI agent guardrails and mTLS support for O11 APIs, and published a blog post citing new Gartner research that names Enterprise Low-Code Application Platforms among the ten critical technology markets for operationalizing AI in 2026. Together, they point to the same conclusion: OutSystems is positioning itself, and its partners, around governed, production-ready enterprise AI rather than pilot-stage experimentation.
Custom guardrail policies for AI agents (August 6, 2026). System Administrators can now define organization-wide rules for denied topics, word and phrase filters, and custom PII patterns, applied as a baseline to every AI agent on the platform.
mTLS support for O11 APIs (July 31, 2026). O11 REST APIs can now authenticate calling clients by certificate, not just credentials, a requirement in many regulated and zero-trust environments. Read our full breakdown.
"Operationalizing enterprise AI: Strategic investments for 2026" (August 4, 2026). Citing Gartner's Critical Technology Markets for Software Engineering 2026 report, OutSystems highlights that Enterprise Low-Code Application Platforms are one of ten technology markets Gartner considers critical to scaling AI across the enterprise through 2030. The post frames the platform as the "execution layer" that turns AI strategy into shippable, governed applications. Read our full breakdown.
None of these three updates is flashy on its own. Together, though, they answer the question most IT leaders are now asking: not "can we build an AI agent?" but "can we build one we're allowed to put into production?". Guardrails and mTLS are exactly the kind of unglamorous, compliance-grade plumbing that decides whether an agentic AI project survives a bank's, insurer's or public sector client's security review.
This is also where the Gartner reference matters more than it looks. Gartner isn't naming low-code critical because it generates code faster; it's naming it critical because platforms like OutSystems now bundle the governance, integration and lifecycle controls that make AI safe to operationalize. That's the same shift we help clients navigate every day as a partner: the technical decision (which AI model, which agent framework) is rarely the bottleneck anymore. Governance, data access and legacy integration are.
For companies in regulated sectors such as banking, insurance, healthcare and public administration, mTLS and custom guardrails aren't nice-to-haves. They're often the difference between a security team approving a pilot for production or sending it back for another quarter of review. Having these controls native to the platform, instead of bolted on by each project team, shortens that review cycle considerably.
What are OutSystems' new AI agent guardrails? Custom, organization-wide rules that System Administrators can define to block or mask specific topics, terms or data patterns across every AI agent built on the platform, going beyond OutSystems' default guardrail categories.
What does mTLS support do for O11 applications? It lets O11 REST APIs authenticate calling clients by certificate, not just by credentials, which is required in many regulated and zero-trust environments.
Why did Gartner name low-code platforms a critical technology for AI in 2026? Because Gartner expects the software development life cycle to become a unified, AI-driven workflow by 2030, and platforms like OutSystems provide the integration, governance and lifecycle tooling needed to operationalize AI safely at scale, not just prototype it.
Ready to take an AI agent through your security review? Talk to our team.
Sources - OutSystems Product Updates, "New agent guardrail policies," August 6, 2026 - OutSystems Product Updates, "mTLS support for API endpoints in O11," July 31, 2026 - OutSystems Blog, "Operationalizing enterprise AI: Strategic investments for 2026," August 4, 2026 - Gartner, Critical Technology Markets for Software Engineering 2026, July 6, 2026 (cited in the OutSystems blog post above)