<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Useful Machines</title>
    <link>https://usefulmachines.co.uk</link>
    <description>Practical AI, useful systems, better workflows. Less hype, more leverage.</description>
    <language>en</language>
    <lastBuildDate>Wed, 20 May 2026 09:06:54 GMT</lastBuildDate>
    <atom:link href="https://usefulmachines.co.uk/rss.xml" rel="self" type="application/rss+xml" />
    <item>
      <title><![CDATA[The Autonomy Ladder]]></title>
      <link>https://usefulmachines.co.uk/writing/the-autonomy-ladder</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/the-autonomy-ladder</guid>
      <description><![CDATA[Google's A2A protocol and MCP are live under shared governance. The protocol layer is solved. The trust layer that determines whether agents can actually discover and rely on each other is not.]]></description>
      <pubDate>Thu, 16 Apr 2026 19:54:19 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[agents]]></category>
      <category><![CDATA[mcp]]></category>
      <category><![CDATA[protocols]]></category>
      <category><![CDATA[security]]></category>
    </item>
  <item>
      <title><![CDATA[Queryable Thinking]]></title>
      <link>https://usefulmachines.co.uk/writing/queryable-thinking</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/queryable-thinking</guid>
      <description><![CDATA[What happens when your writing becomes a callable knowledge service and the reader becomes a synthesiser.]]></description>
      <pubDate>Thu, 16 Apr 2026 18:18:48 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[mcp]]></category>
      <category><![CDATA[content-architecture]]></category>
      <category><![CDATA[writing]]></category>
      <category><![CDATA[memory]]></category>
    </item>
  <item>
      <title><![CDATA[The Productivity Paradox: Why AI Is Making Us More Burnt Out, Not Less]]></title>
      <link>https://usefulmachines.co.uk/writing/productivity-paradox-ai-burnout</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/productivity-paradox-ai-burnout</guid>
      <description><![CDATA[AI collapsed the natural bottlenecks in knowledge work, and those bottlenecks were doing more than we realised. Why efficiency gains are producing burnout, not breathing room.]]></description>
      <pubDate>Wed, 01 Apr 2026 16:10:57 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[productivity]]></category>
      <category><![CDATA[burnout]]></category>
      <category><![CDATA[adoption]]></category>
      <category><![CDATA[work]]></category>
    </item>
  <item>
      <title><![CDATA[Inside the Memory Layer]]></title>
      <link>https://usefulmachines.co.uk/writing/inside-the-memory-layer</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/inside-the-memory-layer</guid>
      <description><![CDATA[How I built a persistent AI memory layer — MCP server, control panel, embeddings, and a query UI that turns conversation history into usable knowledge.]]></description>
      <pubDate>Wed, 01 Apr 2026 06:40:10 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[memory]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[tools]]></category>
    </item>
  <item>
      <title><![CDATA[The memory layer is the moat]]></title>
      <link>https://usefulmachines.co.uk/writing/solving-the-memory-problem</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/solving-the-memory-problem</guid>
      <description><![CDATA[I built a persistent memory layer that sits outside any single AI provider. Now every tool I use shares the same context, and it turns out the memory matters more than the model.]]></description>
      <pubDate>Wed, 01 Apr 2026 06:38:35 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[memory]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[tools]]></category>
    </item>
  <item>
      <title><![CDATA[Prompting Is Not the Moat]]></title>
      <link>https://usefulmachines.co.uk/writing/prompting-is-not-the-moat</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/prompting-is-not-the-moat</guid>
      <description><![CDATA[Prompt engineering courses are booming, but the skill they teach is depreciating fast. The durable value in AI sits above the interface.]]></description>
      <pubDate>Mon, 30 Mar 2026 19:19:17 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[tools]]></category>
      <category><![CDATA[workflows]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[adoption]]></category>
    </item>
  <item>
      <title><![CDATA[The Alignment Problem Is a Human Problem]]></title>
      <link>https://usefulmachines.co.uk/writing/the-alignment-problem-is-a-human-problem</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/the-alignment-problem-is-a-human-problem</guid>
      <description><![CDATA[AI alignment research assumes humans can specify what they want. Behavioural science says otherwise.]]></description>
      <pubDate>Mon, 30 Mar 2026 08:22:15 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[alignment]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[adoption]]></category>
    </item>
  <item>
      <title><![CDATA[AI feels new. The labour pattern isn't.]]></title>
      <link>https://usefulmachines.co.uk/writing/ai-feels-new-the-labour-pattern-isnt</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/ai-feels-new-the-labour-pattern-isnt</guid>
      <description><![CDATA[Every few decades, a technology arrives that triggers the same prediction - this one will finally make people obsolete. The technology changes, but the mistake doesn't.]]></description>
      <pubDate>Mon, 30 Mar 2026 08:21:16 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[labour]]></category>
      <category><![CDATA[future-of-work]]></category>
    </item>
  <item>
      <title><![CDATA[What OpenClaw Actually Proved]]></title>
      <link>https://usefulmachines.co.uk/writing/what-openclaw-actually-proved-v2</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/what-openclaw-actually-proved-v2</guid>
      <description><![CDATA[Jensen Huang ranked OpenClaw alongside Linux. The real lesson is that users wanted AI with tool access and execution rights, and the industry spent three years optimising for the wrong thing.]]></description>
      <pubDate>Mon, 30 Mar 2026 08:19:24 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[agents]]></category>
      <category><![CDATA[systems]]></category>
      <category><![CDATA[tools]]></category>
    </item>
  <item>
      <title><![CDATA[AI Is Making Up Its Own Values]]></title>
      <link>https://usefulmachines.co.uk/writing/ai-is-making-up-its-own-values</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/ai-is-making-up-its-own-values</guid>
      <description><![CDATA[AI models are developing coherent internal value systems. Some of those values are ones we wouldn't choose.]]></description>
      <pubDate>Mon, 30 Mar 2026 08:18:13 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[alignment]]></category>
      <category><![CDATA[safety]]></category>
      <category><![CDATA[values]]></category>
      <category><![CDATA[misalignment]]></category>
    </item>
  <item>
      <title><![CDATA[DeepSeek thinks it's Claude]]></title>
      <link>https://usefulmachines.co.uk/writing/deepseek-thinks-its-claude</link>
      <guid isPermaLink="true">https://usefulmachines.co.uk/writing/deepseek-thinks-its-claude</guid>
      <description><![CDATA[An AI that doesn't know who it is turned out to be a fingerprint of industrial-scale model distillation - and the ethics are more complicated than anyone wants to admit]]></description>
      <pubDate>Sat, 28 Mar 2026 16:50:47 GMT</pubDate>
      <category><![CDATA[ai]]></category>
      <category><![CDATA[distillation]]></category>
      <category><![CDATA[identity]]></category>
      <category><![CDATA[deepseek]]></category>
      <category><![CDATA[anthropic]]></category>
    </item>
  </channel>
</rss>