<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Dtwo Blog</title><description>Notes from the team building Dtwo: product updates, engineering deep dives, culture, and the start-up journey, written as it happens.</description><link>https://blog.dtwo.ai/</link><item><title>AI TRiSM for AI Agents: One Policy, Every Agent</title><link>https://blog.dtwo.ai/posts/ai-trism-for-ai-agents/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/ai-trism-for-ai-agents/</guid><description>AI TRiSM is now a real budget line. Centralizing AI agent policy means one set of policies, enforced on every agent, with one audit trail.</description><pubDate>Mon, 16 Nov 2026 09:00:00 GMT</pubDate></item><item><title>How to Let Claude Code Run Without Permission Prompts, Safely</title><link>https://blog.dtwo.ai/posts/let-claude-code-run-without-permission-prompts/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/let-claude-code-run-without-permission-prompts/</guid><description>Permission prompts slow developers down until they turn them off. Here&apos;s how to move those decisions to one governed control point, and how to connect Claude Code, Cursor, Codex or Claude Desktop to it.</description><pubDate>Mon, 09 Nov 2026 09:00:00 GMT</pubDate></item><item><title>Adaptive Policies</title><link>https://blog.dtwo.ai/posts/adaptive-policies/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/adaptive-policies/</guid><description>A policy written once is wrong by next quarter. Here is the design we are building toward: policies that tighten and widen from how your workforce actually uses AI, from real-time signals like a new CVE, with a human accepting every change.</description><pubDate>Wed, 28 Oct 2026 09:00:00 GMT</pubDate></item><item><title>Why Agentic AI Needs Dynamic Safeguards</title><link>https://blog.dtwo.ai/posts/why-agentic-ai-needs-dynamic-safeguards/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/why-agentic-ai-needs-dynamic-safeguards/</guid><description>Static permissions judge each AI agent action alone. Dynamic safeguards react to what the agent already did in the session, so you can say yes to agents on sensitive data.</description><pubDate>Tue, 27 Oct 2026 09:00:00 GMT</pubDate></item><item><title>How we are building our AI GTM System at Dtwo</title><link>https://blog.dtwo.ai/posts/building-an-ai-gtm-system/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/building-an-ai-gtm-system/</guid><description>How Dtwo is building an AI GTM system where agents do real work across the stack, and policy decides what they&apos;re allowed to do.</description><pubDate>Thu, 15 Oct 2026 19:00:00 GMT</pubDate></item><item><title>AI Blast Radius</title><link>https://blog.dtwo.ai/posts/ai-blast-radius/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/ai-blast-radius/</guid><description>Of all the questions to ask before handing an AI agent access to your systems, one does most of the work: if this goes wrong, what is the worst that can happen? Four versions of that question, and how to act on the answers.</description><pubDate>Fri, 02 Oct 2026 01:30:00 GMT</pubDate></item><item><title>NVIDIA OpenShell governs reachability. Dtwo governs meaning.</title><link>https://blog.dtwo.ai/posts/nvidia-openshell-and-dtwo/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/nvidia-openshell-and-dtwo/</guid><description>NVIDIA&apos;s OpenShell, announced September 28, decides what an agent can reach. Dtwo sits in its middleware hook and judges each call by its arguments and by what the session already did elsewhere. If you run agents in OpenShell, you still need that layer.</description><pubDate>Thu, 01 Oct 2026 09:00:00 GMT</pubDate></item><item><title>AI governance evaluation</title><link>https://blog.dtwo.ai/posts/ai-governance-evaluation/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/ai-governance-evaluation/</guid><description>Expressiveness and usability. Everything else on the vendor&apos;s feature list is one of those two in disguise.</description><pubDate>Tue, 29 Sep 2026 09:00:00 GMT</pubDate></item><item><title>I sell agent security. So I gave my agents a boss.</title><link>https://blog.dtwo.ai/posts/i-gave-my-agents-a-boss/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/i-gave-my-agents-a-boss/</guid><description>Our founding AE runs his entire sales job through Dtwo — prospecting, CRM, call prep, follow-up — and explains what a single control point actually buys you.</description><pubDate>Mon, 28 Sep 2026 09:00:00 GMT</pubDate></item><item><title>The Wrong Chris Problem: Why AI Agents Need Smarter Guardrails</title><link>https://blog.dtwo.ai/posts/the-wrong-chris-problem/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/the-wrong-chris-problem/</guid><description>There are two Chrises in our Slack — a coworker and a customer. Here&apos;s the policy I wrote so an agent picking the wrong one can&apos;t turn into a data breach, and what I learned when a policy went too far the other way.</description><pubDate>Sun, 27 Sep 2026 09:00:00 GMT</pubDate></item><item><title>Connecting Claude to Microsoft 365, Safely</title><link>https://blog.dtwo.ai/posts/connecting-claude-to-microsoft-365-safely/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/connecting-claude-to-microsoft-365-safely/</guid><description>Security teams keep saying no when users ask to connect Claude to email. The problem isn&apos;t access, it&apos;s granularity. Here&apos;s what a yes can look like.</description><pubDate>Mon, 14 Sep 2026 09:00:00 GMT</pubDate></item><item><title>My Claude Guardrails</title><link>https://blog.dtwo.ai/posts/my-claude-guardrails/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/my-claude-guardrails/</guid><description>Claude Code can access our internal systems. Here are the guardrails: the policies that decide what my AI agent can actually reach, all published in our policy catalog.</description><pubDate>Sun, 30 Aug 2026 09:00:00 GMT</pubDate></item><item><title>Code got cheap. Architecture got iterative.</title><link>https://blog.dtwo.ai/posts/code-got-cheap-architecture-got-iterative/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/code-got-cheap-architecture-got-iterative/</guid><description>Cheap implementation moved the work upstream. Every architectural decision is now recorded in the same repo as code. It becomes the context for everything built after it, and gets revised when we turn out to be wrong.</description><pubDate>Sat, 15 Aug 2026 09:00:00 GMT</pubDate></item><item><title>How CarNow Drove AI Adoption and Developer Productivity with Dtwo</title><link>https://blog.dtwo.ai/posts/how-carnow-drove-ai-adoption-with-dtwo/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/how-carnow-drove-ai-adoption-with-dtwo/</guid><description>CarNow consolidated access into a single Dtwo gateway to drive AI adoption and measurable productivity gains across its engineering team.</description><pubDate>Fri, 07 Aug 2026 09:00:00 GMT</pubDate></item><item><title>Good policy is the hard part. So we&apos;re sharing ours.</title><link>https://blog.dtwo.ai/posts/good-policy-is-the-hard-part/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/good-policy-is-the-hard-part/</guid><description>The risk from AI agents isn&apos;t any one system — it&apos;s the path between them. Policy guards that path. It&apos;s hard to write — so we open-sourced a library of policies.</description><pubDate>Wed, 01 Jul 2026 09:00:00 GMT</pubDate></item><item><title>Audit kept us honest. Agents don&apos;t care.</title><link>https://blog.dtwo.ai/posts/why-audit-fails-for-ai-agents/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/why-audit-fails-for-ai-agents/</guid><description>Audit worked because people feared getting caught. AI agents don&apos;t — so agent security must shift from audit to authorization across every system.</description><pubDate>Thu, 25 Jun 2026 09:00:00 GMT</pubDate></item><item><title>Securing Claude Cowork</title><link>https://blog.dtwo.ai/posts/securing-claude-cowork/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/securing-claude-cowork/</guid><description>A practical guide to rolling out Claude Cowork safely: restrict risky access paths, route MCP through a policy layer, and monitor what your agents actually do.</description><pubDate>Fri, 12 Jun 2026 12:00:00 GMT</pubDate></item><item><title>Agentic AI security: why legacy controls fail AI agents</title><link>https://blog.dtwo.ai/posts/why-legacy-security-controls-fail-ai-agents/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/why-legacy-security-controls-fail-ai-agents/</guid><description>Legacy security controls were built for humans, not AI agents — that&apos;s why agentic AI security and governance are breaking. Here&apos;s the gap, and how to close it.</description><pubDate>Thu, 11 Jun 2026 09:00:00 GMT</pubDate></item><item><title>Intent-Based Policies for AI</title><link>https://blog.dtwo.ai/posts/intent-based-policies-for-ai/</link><guid isPermaLink="true">https://blog.dtwo.ai/posts/intent-based-policies-for-ai/</guid><description>The policy model that closes the gap between what organizations mean and what their controls enforce.</description><pubDate>Wed, 10 Jun 2026 05:00:00 GMT</pubDate></item></channel></rss>