<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
  <title>Mauro Cofelice’s notebook</title>
  <link>https://maurocofelice.com/en/notebook</link>
  <atom:link href="https://maurocofelice.com/en/feed.xml" rel="self" type="application/rss+xml"/>
  <description>Notes and open questions on artificial intelligence, healthcare and organisational decisions.</description>
  <language>en</language>
  <lastBuildDate>Wed, 11 Nov 2026 08:00:00 +0000</lastBuildDate>
  <item>
    <title>A complex idea doesn’t fit into ten seconds</title>
    <link>https://maurocofelice.com/en/notes/elevator-pitch</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/elevator-pitch</guid>
    <pubDate>Wed, 11 Nov 2026 08:00:00 +0000</pubDate>
    <category>Complex systems</category>
    <description>Let’s go personal today. The 10-second elevator pitch is one of the most overrated (stupid) concepts in the startup ecosystem. And possibly one of the most damaging.</description>
  </item>
  <item>
    <title>AI returns depend on processes</title>
    <link>https://maurocofelice.com/en/notes/ritorno-ai-processi</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/ritorno-ai-processi</guid>
    <pubDate>Tue, 27 Oct 2026 08:00:00 +0000</pubDate>
    <category>AI and governance</category>
    <description>According to McKinsey &amp; Company 80% of people say AI makes them more productive. Only 37% of companies see any EBIT impact, the same share as a year ago. That is the central finding of McKinsey’s State of AI in 2026. So where does the value get stuck? Between the individual and the workflow.</description>
  </item>
  <item>
    <title>Value moves from code to domain</title>
    <link>https://maurocofelice.com/en/notes/valore-codice-dominio</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/valore-codice-dominio</guid>
    <pubDate>Mon, 12 Oct 2026 08:00:00 +0000</pubDate>
    <category>AI and governance</category>
    <description>Fact: AI is commoditizing software. A project like ArtCraft raises a question every software company and system integrator should ask: what happens when software is no longer the primary source of competitive advantage?</description>
  </item>
  <item>
    <title>Time saved goes to the patient, not to volume</title>
    <link>https://maurocofelice.com/en/notes/scribe-ambientale</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/scribe-ambientale</guid>
    <pubDate>Mon, 28 Sep 2026 08:00:00 +0000</pubDate>
    <category>Digital health</category>
    <description>A time-motion study just published in JMIR Medical Informatics followed nine clinicians at a Singapore academic medical center, with and without an ambient AI scribe. Documentation time per visit dropped 15 percent. Eye contact with patients rose almost 11 percent. Total visit duration did not change. That last number is the one vendors rarely put on a slide. The pitch behind most ambient scribe…</description>
  </item>
  <item>
    <title>Who decides when an AI agent decides?</title>
    <link>https://maurocofelice.com/en/notes/chi-decide-agente-ai</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/chi-decide-agente-ai</guid>
    <pubDate>Thu, 06 Aug 2026 08:00:00 +0000</pubDate>
    <category>AI and governance</category>
    <description>In the US the debate on agentic AI in healthcare has moved past “if”. One large insurer now runs tens of thousands of prior authorization decisions a day, most of them handled autonomously. That number is impressive and uncomfortable at the same time. When an agent can act across an EMR, a claims platform and a scheduling system without a human in every step, the real design problem is no longer…</description>
  </item>
  <item>
    <title>The value of AI also lies in the when</title>
    <link>https://maurocofelice.com/en/notes/valore-ia-quando</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/valore-ia-quando</guid>
    <pubDate>Wed, 22 Jul 2026 08:00:00 +0000</pubDate>
    <category>AI and governance</category>
    <description>A university student, Risha Panigrahi, just removed the biggest edge politicians have ever had: the gap between the lie and the fact-check. She built InTruth, a Chrome extension that listens to any live speech, debate, or interview. It transcribes every word and checks each claim in real time, before the speaker even finishes the sentence.</description>
  </item>
  <item>
    <title>AI reduces clinical risk only when it is governed</title>
    <link>https://maurocofelice.com/en/notes/ia-rischio-clinico</link>
    <guid isPermaLink="true">https://maurocofelice.com/en/notes/ia-rischio-clinico</guid>
    <pubDate>Tue, 21 Jul 2026 08:00:00 +0000</pubDate>
    <category>Digital health</category>
    <description>ECRI put AI at the top of its health technology hazards list for 2025. A few months later the Joint Commission and CHAI released their first guidance on responsible AI use in hospitals. Both point at the same gap: deployment is running ahead of governance. That gap is where clinical risk will play out over the next few years.</description>
  </item>
</channel>
</rss>
