trail's AI Governance Education Hub & News Room

Latest posts

CategoriesResponsible AIEU AI ActISO/IEC 42001Press ReleaseCase StudyTechProduct

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.

The Digital Omnibus: what changed in the EU AI Act

The EU AI Act has been amended. Regulation (EU) 2026/1744, the Digital Omnibus on AI, entered into force on 27 July 2026 and moved the deadlines for high-risk AI systems, narrowed the scope, and cut documentation requirements for smaller companies. This article gives a high-level overview of what changed, what stayed the same, and what the new timeline means for your planning.

Tech

AI Agent Governance: Runtime Enforcement with trail

trail's AI Governance suite now includes agent runtime enforcement: configurable permissions on actions agents take using MCPs. All agent calls from the organization are routed through one central gateway, with permissions enforced on each tool call. This helps safeguard agent sessions against prompt injection, unintended exposure of PII, and agents taking actions beyond the scope of their workflow. MCPs can be imported into a centralized registry within trail, permissions configured, then published to the gateway to be enforced.

Responsible AI

The State of AI Governance in 2026: From Compliance Checklists to Continuous Control

A dive into the AI governance landscape - what enterprises need to know when building, buying, or using AI.

Product

Assessing Control Effectiveness at Scale

Control assessments are the core of effective GRC: they show whether risk mitigation measures for your assets are actually working and whether compliance requirements are being fulfilled. trail supports this through efficient automations, all based on actual and current evidence.

Product

Managing Novel AI Risks

AI risk management is the foundation of every mature IT and AI governance program. AI adds a new layer on top of existing risk processes: novel risks, compounded existing ones, and a pace of change no quarterly review can keep up with. trail gives you the library, structure, and automation to manage AI risk continuously.

Product

Govern Third-Party AI Without Slowing Down Innovation

Third-party AI can be found in every organization, as most are not developing their own AI systems but are rather opting for buying and deploying external ones. While this is a good way to quickly make use of specialized vertical AI solutions, this is a big disadvantage from an AI and IT governance perspective, as you’re usually missing insight into the inner workings of those solutions. Additionally, vendors of traditional software introduce new AI-based capabilities, letting your backlog of reassessments of your IT assets grow each week. This article looks at how you can manage these new challenges and how trail can help you solve these.

Product

Automate IT and AI Compliance with Agents

trail’s AI agents can automate whole GRC, IT governance and AI governance processes – with trail, one person can now do the compliance work of five. No matter if you want to automate asset intake and risk categorization, fill out questionnaires, find evidence, or assess controls. Simply bring your process and sources, and let your agent do the work.

Product

The AI Registry with Superpowers

A centralized AI Registry is the foundation for effective AI governance. It provides a living system of record to gain visibility into your AI landscape – from discovering AI to managing assets across the whole lifecycle – while empowering teams to scale compliant AI deployment through automated and self-serve processes.

News

FPT and trail Announce Partnership in AI Governance and Compliance

trail and the global IT services provider FPT have signed a Memorandum of Understanding to jointly advance next-generation AI governance and compliance solutions for enterprises navigating complex regulatory environments.

Tech

Copy-on-Write in Agentic Systems

Copy-on-Write (CoW) enables database-level isolation for agentic workflows, so users can review and approve agent outputs before they permanently affect production data. Agent changes are written to a separate 'changes' table, but appear merged with 'base' (production) data in the agent's view. At the end of a session, the user can selectively commit agent changes to the main database. This article introduces a CoW implementation to a SQL database, as well as a worked example of CoW in a farm Inventory Management System.

1 / 3

Our Blog

The Digital Omnibus: what changed in the EU AI Act

AI Agent Governance: Runtime Enforcement with trail

The State of AI Governance in 2026: From Compliance Checklists to Continuous Control

Assessing Control Effectiveness at Scale

Managing Novel AI Risks

Govern Third-Party AI Without Slowing Down Innovation

Latest posts

The Digital Omnibus: what changed in the EU AI Act

AI Agent Governance: Runtime Enforcement with trail

The State of AI Governance in 2026: From Compliance Checklists to Continuous Control

Assessing Control Effectiveness at Scale

Managing Novel AI Risks

Govern Third-Party AI Without Slowing Down Innovation

Automate IT and AI Compliance with Agents

The AI Registry with Superpowers

FPT and trail Announce Partnership in AI Governance and Compliance

Copy-on-Write in Agentic Systems