<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns:news="http://www.google.com/schemas/sitemap-news/0.9">
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3281-adobe-is-testing-how-well-ai-agents-can-use-editing-tools-like-a-photoshop-profe</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-11-01T14:56:44+00:00</news:publication_date>
      <news:title>Adobe test AI voor professionele beeldbewerking in Photoshop</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3271-ook-grok-gaat-watermerken-toevoegen-aan-tekst</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-11-01T10:07:31+00:00</news:publication_date>
      <news:title>Grok voegt onzichtbare watermerken toe aan tekstoutput</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3269-cvss-10-0-lxd-microcloud-flaw-lets-attackers-impersonate-trusted-clients-without</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-11-01T08:51:31+00:00</news:publication_date>
      <news:title>Kritieke kwetsbaarheid in LXD en MicroCloud stelt aanvallers in staat tot impersonatie</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3265-nace-ai-open-sources-drex-max-v1-a-27b-decision-model-that-runs-locally-on-apple</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-11-01T08:10:07+00:00</news:publication_date>
      <news:title>Nace AI maakt Drex Max v1 open-source voor Apple Silicon</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3264-fbi-links-north-korea-s-tradertraitor-to-record-6-43-billion-binance-crypto-heis</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-11-01T06:44:27+00:00</news:publication_date>
      <news:title>FBI koppelt Noord-Koreaanse groep aan recorddiefstal van $6,43 miljard</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3261-how-we-took-over-chatgpt-accounts</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T23:15:22+00:00</news:publication_date>
      <news:title>Serieuze kwetsbaarheid ontdekt in ChatGPT die privégegevens kan lekken</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3257-anthropic-s-watermark-rollout-leaves-a-loophole-on-two-older-claude-models</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T19:20:55+00:00</news:publication_date>
      <news:title>Anthropic breidt tekstwatermerk uit, oudere modellen kwetsbaar</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3256-northflank-security-breach-leads-to-abuse-of-customers-ai-api-keys</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T18:01:12+00:00</news:publication_date>
      <news:title>Northflank ontdekt beveiligingslek bij klanten met AI API-sleutels</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3251-reddit-to-lock-content-behind-logins-to-fight-scrapers</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T14:05:15+00:00</news:publication_date>
      <news:title>Reddit plaatst content achter inlog om scrapers te bestrijden</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3247-meet-factory-droid-1-0-a-model-independent-coding-agent-harness-built-on-script</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T10:38:14+00:00</news:publication_date>
      <news:title>Factory lanceert Droid 1.0, model-onafhankelijke coding agent</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3241-openai-raises-just-11-billion-of-its-30-billion-target-as-big-investors-lose-the</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T09:22:52+00:00</news:publication_date>
      <news:title>OpenAI haalt $11 miljard op van $30 miljard doelstelling</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3240-attackers-chain-three-ivanti-connect-secure-flaws-to-gain-unauthenticated-root-a</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T08:27:43+00:00</news:publication_date>
      <news:title>Aanvallers misbruiken drie kwetsbaarheden in Ivanti Connect Secure</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://redflare.nl/ai-nieuws/artikel.php?a=3237-a-shared-chatgpt-link-could-run-attacker-code-in-your-session-and-read-private-c</loc>
    <news:news>
      <news:publication><news:name>RedFlare AI-nieuws</news:name><news:language>nl</news:language></news:publication>
      <news:publication_date>2026-10-31T05:33:41+00:00</news:publication_date>
      <news:title>Kwetsbaarheid in ChatGPT laat aanvallers privégesprekken lezen</news:title>
    </news:news>
  </url>
</urlset>
