<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>PAICE Assessment Changelog</title>
    <link>https://paice.work/changelog.xml</link>
    <atom:link href="https://paice.work/changelog.xml" rel="self" type="application/rss+xml" />
    <description>User-facing changes to the PAICE assessment, scoring, and platform. Scoped to changes that materially affect the assessment experience or interpretation. Backend implementation details, infrastructure, and security-sensitive configuration are intentionally excluded.</description>
    <language>en</language>
    <copyright>PAICE.work PBC</copyright>
    <managingEditor>success@paice.work (PAICE.work PBC)</managingEditor>
    <lastBuildDate>Wed, 27 May 2026 00:00:00 GMT</lastBuildDate>
    <ttl>1440</ttl>

    <item>
      <title>Site repositioned around operational AI risk visibility</title>
      <link>https://paice.work/</link>
      <guid isPermaLink="false">paice-positioning-2026-05-26</guid>
      <pubDate>Tue, 26 May 2026 00:00:00 GMT</pubDate>
      <category>positioning</category>
      <description><![CDATA[
        Public-facing surfaces (homepage, About, FAQ, Baseline) rewritten to lead with operational AI risk visibility rather than AI-collaboration education. New FAQ category "Why Existing Measurement Systems Fail." Drift-sequence visual added to the Baseline page. Scoring methodology and the assessment itself are unchanged.
      ]]></description>
    </item>

    <item>
      <title>PAICE Pulse launched (3-minute confidence check)</title>
      <link>https://paice.work/pulse</link>
      <guid isPermaLink="false">pulse-launch-2026-04</guid>
      <pubDate>Mon, 14 Apr 2026 00:00:00 GMT</pubDate>
      <category>product</category>
      <description><![CDATA[
        PAICE Pulse went live as a 3-minute self-identification on five archetypes (Delegator, Checker, Strategist, Skeptic, Adopter). Pulse does not score behavioral reliability; it identifies collaboration style. The 25-minute behavioral assessment remains the source of the PAICE Score.
      ]]></description>
    </item>

    <item>
      <title>AI Capability Baseline pricing published</title>
      <link>https://paice.work/baseline</link>
      <guid isPermaLink="false">baseline-pricing-2026-04</guid>
      <pubDate>Tue, 01 Apr 2026 00:00:00 GMT</pubDate>
      <category>product</category>
      <description><![CDATA[
        Cohort engagement tiers published with flat-fee pricing: Team ($3,500, 12-20 people, 2 weeks), Baseline ($9,900, 21-50 people, 4 weeks), Enterprise ($19,500, 51-100 people). All tiers operate without integrations, system access, or personal data collection.
      ]]></description>
    </item>

    <item>
      <title>Spanish localization released</title>
      <link>https://paice.work/</link>
      <guid isPermaLink="false">i18n-spanish-2026-03</guid>
      <pubDate>Mon, 31 Mar 2026 00:00:00 GMT</pubDate>
      <category>internationalization</category>
      <description><![CDATA[
        Spanish-language assessment and site interface released. First non-English locale; established the localization pipeline used for subsequent languages. French, Portuguese, and Urdu followed in Q2.
      ]]></description>
    </item>

    <item>
      <title>Token formats migrated to HardGuard25 alphabet</title>
      <link>https://hardguard25.com/</link>
      <guid isPermaLink="false">hardguard25-migration-2026-03</guid>
      <pubDate>Fri, 06 Mar 2026 00:00:00 GMT</pubDate>
      <category>security</category>
      <description><![CDATA[
        All access codes and tokens shown to users were migrated to the HardGuard25 25-character visually unambiguous alphabet (https://hardguard25.com/). Removes confusables (B/8, E/3, I/1, L/1, O/0, Q/O, S/5, T/+, V/U, X/*, Z/2) plus dyslexia mirror pairs. Existing tokens continue to validate.
      ]]></description>
    </item>

    <item>
      <title>TEE Confidential Mode scoped to paid tiers</title>
      <link>https://paice.work/whitepaper</link>
      <guid isPermaLink="false">tee-confidential-paid-2026-03</guid>
      <pubDate>Tue, 04 Mar 2026 00:00:00 GMT</pubDate>
      <category>privacy</category>
      <description><![CDATA[
        Cryptographically-verifiable confidential assessment mode (using Trusted Execution Environments) moved to optional availability under PAICE Cohort and paid tiers once launched. Standard mode remains the default for free individual assessments and prioritizes scoring quality.
      ]]></description>
    </item>

    <item>
      <title>Dual-mode architecture: Standard vs. TEE Confidential</title>
      <link>https://paice.work/whitepaper</link>
      <guid isPermaLink="false">dual-mode-launch-2026-02</guid>
      <pubDate>Thu, 19 Feb 2026 00:00:00 GMT</pubDate>
      <category>privacy</category>
      <description><![CDATA[
        Launched dual-mode architecture: Standard mode optimizes scoring quality; TEE Confidential mode offers mathematically verifiable privacy guarantees via cryptographically attested execution. Documented in the privacy/security/compliance whitepaper (NEARCon edition).
      ]]></description>
    </item>

    <item>
      <title>Cohort assessment infrastructure launched</title>
      <link>https://paice.work/baseline</link>
      <guid isPermaLink="false">cohort-v1-launch-2026-02</guid>
      <pubDate>Mon, 17 Feb 2026 00:00:00 GMT</pubDate>
      <category>product</category>
      <description><![CDATA[
        Cohort assessment infrastructure went live: participant tokens, cohort creation, basic participant flows. Initial scope: 20-100 person cohorts. Reporting is strictly cohort-level; individual scores cannot be disclosed to enterprise buyers by architectural design.
      ]]></description>
    </item>

    <item>
      <title>Scoring scale migrated from 0-100 to 0-1000</title>
      <link>https://paice.work/faq</link>
      <guid isPermaLink="false">scoring-0-1000-2026-02</guid>
      <pubDate>Wed, 12 Feb 2026 00:00:00 GMT</pubDate>
      <category>scoring</category>
      <description><![CDATA[
        Canonical PAICE Score scale changed from 0-100 to 0-1000, with FICO-style tier names (Constrained, Informed, Proficient, Advanced, Exceptional). The 0-1000 scale is intentionally not a percentage and does not represent "% of the way to mastery." This change was made to avoid grade-like interpretations that triggered shame responses in early users.
      ]]></description>
    </item>

  </channel>
</rss>
