<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel>
  <title>veli(Q Lab Insights</title>
  <link>https://veliqlab.com/blog</link>
  <atom:link href="https://veliqlab.com/feed.xml" rel="self" type="application/rss+xml"/>
  <description>Notes from the lab on autonomy, governance and the road to the autonomous organization.</description>
  <language>en</language>
  <item><title>How to choose the first function to automate</title><link>https://veliqlab.com/blog/choosing-the-first-function-to-automate</link><guid>https://veliqlab.com/blog/choosing-the-first-function-to-automate</guid><pubDate>Fri, 16 Oct 2026 09:00:00 GMT</pubDate><category>Playbook</category><description>The first function you give to an autonomous system decides how much trust, budget and momentum you have afterwards. A practical way to choose it.</description></item>
  <item><title>Autonomy is a dial, not a switch</title><link>https://veliqlab.com/blog/autonomy-is-a-dial-not-a-switch</link><guid>https://veliqlab.com/blog/autonomy-is-a-dial-not-a-switch</guid><pubDate>Fri, 09 Oct 2026 09:00:00 GMT</pubDate><category>Thinking</category><description>Most conversations about AI ask whether to automate something. The better question is how much authority to grant, and how that authority is earned.</description></item>
  <item><title>Agentic AI vs AI agents: what is the difference?</title><link>https://veliqlab.com/blog/agentic-ai-vs-ai-agents</link><guid>https://veliqlab.com/blog/agentic-ai-vs-ai-agents</guid><pubDate>Fri, 09 Oct 2026 09:00:00 GMT</pubDate><category>Thinking</category><description>The two terms are used interchangeably, and that causes confusion when you buy or build. A plain-English comparison, with examples and a simple test.</description></item>
  <item><title>Why most agentic AI projects stall, and how to avoid it</title><link>https://veliqlab.com/blog/why-agentic-ai-projects-stall</link><guid>https://veliqlab.com/blog/why-agentic-ai-projects-stall</guid><pubDate>Tue, 06 Oct 2026 09:00:00 GMT</pubDate><category>Strategy</category><description>Gartner expects more than 40% of agentic AI projects to be cancelled by the end of 2027. The reasons are predictable, which means they are avoidable.</description></item>
  <item><title>A day in an autonomous finance function</title><link>https://veliqlab.com/blog/a-day-in-an-autonomous-finance-function</link><guid>https://veliqlab.com/blog/a-day-in-an-autonomous-finance-function</guid><pubDate>Fri, 02 Oct 2026 09:00:00 GMT</pubDate><category>Playbook</category><description>What does it look like when reconciliation, anomaly detection and the close run continuously, and people hold the final signature? An illustrative day, hour by hour.</description></item>
  <item><title>What comes after agents: the orchestrated enterprise</title><link>https://veliqlab.com/blog/what-comes-after-agents</link><guid>https://veliqlab.com/blog/what-comes-after-agents</guid><pubDate>Tue, 29 Sep 2026 09:00:00 GMT</pubDate><category>Frontier</category><description>Autonomous functions are the first step. The next is functions that share context and act together. A look at where the technology is heading and what it will require.</description></item>
  <item><title>Designing AI for Thailand: language, channels and data protection</title><link>https://veliqlab.com/blog/designing-ai-for-thailand</link><guid>https://veliqlab.com/blog/designing-ai-for-thailand</guid><pubDate>Thu, 24 Sep 2026 09:00:00 GMT</pubDate><category>Regional</category><description>AI built for a global default often disappoints in Southeast Asia. Four practical lessons from building for customers who live on LINE, write in Thai and expect an instant reply.</description></item>
</channel></rss>
