<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>Fireworks Ai | FirewallSync</title><link>https://firewallsync.com/tags/fireworks-ai/</link><description>Practical cybersecurity writing for engineering teams that ship fast.</description><language>en-us</language><image><url>https://firewallsync.com/icon-512.png</url><title>FirewallSync</title><link>https://firewallsync.com/</link></image><lastBuildDate>Sat, 14 Nov 2026 04:50:00 +0000</lastBuildDate><atom:link href="https://firewallsync.com/tags/fireworks-ai/index.xml" rel="self" type="application/rss+xml"/><item><title>Uber Driver License Data Stolen in Fireworks AI Breach</title><link>https://firewallsync.com/posts/uber-driver-license-data-stolen-fireworks-ai-breach-what-we-know/</link><guid>https://firewallsync.com/posts/uber-driver-license-data-stolen-fireworks-ai-breach-what-we-know/</guid><pubDate>Sat, 14 Nov 2026 04:50:00 +0000</pubDate><dc:creator>Craig Leslie</dc:creator><category>AI Security</category><description>Uber says an unauthorized party used compromised credentials at a service provider to reach Uber data, and TechCrunch reports the provider is Fireworks AI and that drivers' license information was taken. State filings so far list 7,870 people in three states. What is confirmed, the timeline, and what engineering teams should check.</description><enclosure url="https://firewallsync.com/posts/uber-driver-license-data-stolen-fireworks-ai-breach-what-we-know/cover_hu3462212548839934181.jpg" type="image/jpeg" length="0"/><media:content url="https://firewallsync.com/posts/uber-driver-license-data-stolen-fireworks-ai-breach-what-we-know/cover_hu3462212548839934181.jpg" medium="image" type="image/jpeg" width="1200" height="630"/><content:encoded><![CDATA[<p><img src="https://firewallsync.com/posts/uber-driver-license-data-stolen-fireworks-ai-breach-what-we-know/cover_hu3462212548839934181.jpg" alt="A dark car with red tail lights driving along a wet city road at night, with the background streaked by motion blur" width="1200" height="630"></p><p><strong>What happened:</strong> Uber says an unauthorized party used compromised credentials at one of its service providers to access Uber data. <a href="https://techcrunch.com/2026/11/13/uber-confirms-hackers-stole-thousands-of-drivers-license-information-in-data-breach/">TechCrunch</a> reports that the provider is <strong>Fireworks AI</strong>, the AI inference company whose breach <a href="/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/">exposed Notion AI chat transcripts</a>, and that drivers&rsquo; personal and driver&rsquo;s license information was stolen. State filings reported so far list 7,870 people across three states; Uber has not given a total.</p>
<p>Uber is the second well-known company this week to say its data was taken through Fireworks. That makes it a useful case for any team that sends data or credentials to an AI vendor.</p>
<h2 id="what-uber-has-confirmed">What Uber has confirmed</h2>
<p>The core facts come from Uber&rsquo;s own breach notice, as reported by <a href="https://finwire.io/news/stock-markets-news/uber-says-us-drivers-license-data-stolen-in-data-breach">Reuters</a> and TechCrunch on November 13, 2026:</p>
<ul>
<li><strong>Cause.</strong> Per the notice on the California attorney general&rsquo;s website, an &ldquo;unauthorized party used compromised credentials at a service provider to access its data&rdquo; (Reuters&rsquo; paraphrase of the filing).</li>
<li><strong>Vendor.</strong> Uber, in a statement reported by TechCrunch, described &ldquo;a security incident impacting one of our third-party vendors, Fireworks.ai.&rdquo; Reuters says the filing itself did not name the provider.</li>
<li><strong>Data.</strong> TechCrunch reports driver notices say personal and driver&rsquo;s license information was stolen. Reuters notes the California filing did not list the data types.</li>
<li><strong>Response.</strong> Uber severed its data connections with the provider and opened an investigation, and is offering credit monitoring and fraud assistance through Cyberscout, a TransUnion company (Reuters).</li>
<li><strong>Misuse.</strong> Uber said it &ldquo;had not identified any misuse of the data to date&rdquo; (Reuters).</li>
</ul>
<h2 id="timeline">Timeline</h2>
<table>
  <thead>
      <tr>
          <th>Date (2026)</th>
          <th>Event</th>
          <th>Source</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Oct 6</td>
          <td>Fireworks identifies unauthorized use of internal Fireworks credentials</td>
          <td>Fireworks bulletin (see our <a href="/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/">Notion report</a>)</td>
      </tr>
      <tr>
          <td>Oct 9</td>
          <td>Provider tells Uber its credentials were compromised</td>
          <td>Reuters, TechCrunch</td>
      </tr>
      <tr>
          <td>Oct 13</td>
          <td>Uber&rsquo;s investigation determines the credentials were used to access Uber data</td>
          <td>Reuters</td>
      </tr>
      <tr>
          <td>Oct 23-24</td>
          <td>Separate incident at a cloud provider hosting Fireworks R&amp;D machines; unauthorized workload run</td>
          <td>Fireworks update</td>
      </tr>
      <tr>
          <td>Nov 10</td>
          <td>Date on Uber&rsquo;s notice</td>
          <td>Reuters</td>
      </tr>
      <tr>
          <td>Nov 13</td>
          <td>Uber&rsquo;s notices and state filings reported</td>
          <td>TechCrunch, Reuters</td>
      </tr>
  </tbody>
</table>
<h2 id="how-many-people-what-the-filings-show">How many people: what the filings show</h2>
<table>
  <thead>
      <tr>
          <th>State</th>
          <th>People</th>
          <th>Source</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Massachusetts</td>
          <td>2,567</td>
          <td>TechCrunch</td>
      </tr>
      <tr>
          <td>South Carolina</td>
          <td>3,430</td>
          <td>TechCrunch</td>
      </tr>
      <tr>
          <td>Rhode Island</td>
          <td>1,873</td>
          <td>TechCrunch, Reuters</td>
      </tr>
      <tr>
          <td>Illinois</td>
          <td>affected, count pending</td>
          <td>TechCrunch</td>
      </tr>
      <tr>
          <td>California</td>
          <td>drivers notified, no number given</td>
          <td>TechCrunch</td>
      </tr>
  </tbody>
</table>
<p>The three numbered states add up to 7,870. That is a floor for those states, not a total: TechCrunch says the figure is expected to rise, and Reuters reports the notice also carried instructions for residents of the District of Columbia, Maryland, North Carolina, New York and Pennsylvania.</p>
<h2 id="confirmed-vs-not-confirmed">Confirmed vs not confirmed</h2>
<table>
  <thead>
      <tr>
          <th>Claim</th>
          <th>Status</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Unauthorized access to Uber data using credentials compromised at a service provider</td>
          <td>Confirmed by Uber&rsquo;s notice (via Reuters)</td>
      </tr>
      <tr>
          <td>The provider is Fireworks AI</td>
          <td>Confirmed by Uber, as reported by TechCrunch</td>
      </tr>
      <tr>
          <td>Driver&rsquo;s license and personal information stolen</td>
          <td>Reported by TechCrunch from driver notices</td>
      </tr>
      <tr>
          <td>Total number of people affected</td>
          <td>Not disclosed</td>
      </tr>
      <tr>
          <td>Which Fireworks incident path (October 6 credentials or October 23 cloud provider) reached Uber data</td>
          <td>Not stated by Uber or Fireworks</td>
      </tr>
      <tr>
          <td>What Uber data Fireworks held, and why</td>
          <td>Not stated</td>
      </tr>
      <tr>
          <td>Who did it</td>
          <td>Fireworks links its incidents to a financially motivated group selling stolen LLM access; it has not named the group, and TechCrunch says no group has claimed the Uber data theft</td>
      </tr>
  </tbody>
</table>
<p>The dates line up with Fireworks&rsquo; first incident (identified October 6, with Uber told on October 9), but neither company has said so. Treat that link as unconfirmed.</p>
<h2 id="what-this-means-for-engineering-teams">What this means for engineering teams</h2>
<p>Uber&rsquo;s notice describes the failure in one phrase: credentials at a service provider. Fireworks&rsquo; own bulletin, covered in our <a href="/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/">Notion report</a>, said it believes API keys and tokens stored as secrets in some customer accounts&rsquo; evaluation jobs were captured. Whatever the exact path here, the pattern to check in your own environment is the same.</p>
<ol>
<li><strong>List what you have handed to AI vendors.</strong> API keys, cloud storage tokens, database or data-warehouse connectors, webhook secrets and service-account keys placed in vendor job settings, eval configs or &ldquo;bring your own bucket&rdquo; integrations. Most teams track the vendor; few track the secrets inside it.</li>
<li><strong>Scope those credentials to one job.</strong> Read-only, single bucket or dataset, short expiry where the vendor supports it. A key that can read the whole driver table should not sit in an evaluation job.</li>
<li><strong>Rotate at the issuer.</strong> Deleting a secret from the vendor&rsquo;s store does not revoke it. Revoke and reissue in your own cloud or identity provider; our guide on <a href="/posts/how-to-detect-and-rotate-leaked-api-keys-before-attackers-use-them/">detecting and rotating leaked API keys</a> covers the order of operations.</li>
<li><strong>Cut the data path, then investigate.</strong> Uber says it severed its data connections with the provider. Know in advance which network paths, connectors and OAuth grants you would disable, and who owns each.</li>
<li><strong>Watch access logs for the vendor&rsquo;s credentials.</strong> If a vendor-held key was used from unfamiliar infrastructure, your own cloud audit logs are the fastest place to see it.</li>
<li><strong>Ask vendors per-feature questions.</strong> Which features store your data or credentials, where, and for how long. Our <a href="/posts/vetting-third-party-saas-tools-before-they-touch-your-data/">third-party SaaS vetting guide</a> has a question list to start from.</li>
</ol>
<h2 id="limits-of-this-article">Limits of this article</h2>
<p>We could not retrieve the California attorney general filing or the state notices directly in this run; the filing details are as reported by Reuters (via finwire.io syndication) and TechCrunch on November 13, 2026. Fireworks did not respond to either outlet. We will add a dated update if Uber or Fireworks publishes more detail.</p>
<h2 id="sources-and-verification">Sources and verification</h2>
<table>
  <thead>
      <tr>
          <th>Important claim</th>
          <th>Source</th>
          <th>Status</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Notice says compromised credentials at a service provider used to access Uber data; notice dated Nov 10; Oct 9 and Oct 13 dates; Cyberscout; no misuse identified; RI 1,873; DC, MD, NC, NY, PA instructions</td>
          <td><a href="https://finwire.io/news/stock-markets-news/uber-says-us-drivers-license-data-stolen-in-data-breach">Reuters via finwire.io, Nov 13, 2026</a></td>
          <td>Verified (press report of the filing)</td>
      </tr>
      <tr>
          <td>Uber names Fireworks.ai; driver&rsquo;s license data; MA, SC, RI counts; Illinois and California status; no group has claimed it</td>
          <td><a href="https://techcrunch.com/2026/11/13/uber-confirms-hackers-stole-thousands-of-drivers-license-information-in-data-breach/">TechCrunch, Nov 13, 2026</a></td>
          <td>Verified (press report)</td>
      </tr>
      <tr>
          <td>Fireworks&rsquo; October 6 and October 23-24 incidents, LLMJacking description</td>
          <td>Fireworks bulletins, as cited in our <a href="/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/">Notion report</a></td>
          <td>Verified (Nov 13 run)</td>
      </tr>
      <tr>
          <td>7,870 combined across three states</td>
          <td>Sum of the state counts above</td>
          <td>Computed</td>
      </tr>
  </tbody>
</table>
]]></content:encoded></item><item><title>Notion AI Chats Exposed in Fireworks AI Breach: What We Know</title><link>https://firewallsync.com/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/</link><guid>https://firewallsync.com/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/</guid><pubDate>Fri, 13 Nov 2026 04:40:00 +0000</pubDate><dc:creator>Craig Leslie</dc:creator><category>AI Security</category><description>Notion says chat transcripts from fewer than 1% of workspaces were exposed after attackers reached its inference vendor Fireworks AI. What Notion and Fireworks have confirmed, the October timeline, and what to check if your product sends prompts to an AI vendor.</description><enclosure url="https://firewallsync.com/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/cover_hu690302835235202017.jpg" type="image/jpeg" length="0"/><media:content url="https://firewallsync.com/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/cover_hu690302835235202017.jpg" medium="image" type="image/jpeg" width="1200" height="630"/><content:encoded><![CDATA[<p><img src="https://firewallsync.com/posts/notion-ai-chats-exposed-fireworks-ai-breach-what-we-know/cover_hu690302835235202017.jpg" alt="Macro close-up inside a hard drive: an orange flexible circuit with copper contacts curving around the silver spindle hub, next to a teal circuit board" width="1200" height="630"></p><p><strong>What happened.</strong> Notion says a security incident at Fireworks AI, a vendor it used to run open-weight models and classification jobs, exposed Notion AI chat transcripts from fewer than 1% of Notion workspaces. Fireworks says the intruder came in through a third-party cloud provider that hosted its R&amp;D machines, and Notion says it has cut its systems off from Fireworks and is notifying affected customers. SecurityWeek and VentureBeat reported the disclosure on November 11 and 12, 2026.</p>
<p>This post sticks to what Notion and Fireworks have published, attributes everything else, and ends with the checks worth running if your own product sends prompts or customer content to an AI inference vendor.</p>
<h2 id="what-notion-has-confirmed">What Notion has confirmed</h2>
<p>From Notion&rsquo;s help-center notice, <a href="https://notion.com/help/security-incident-involving-a-third-party-ai-vendor">Notice about the Fireworks AI incident that may impact some Notion workspaces</a> (the page carries no publish date; we read it on November 13, 2026):</p>
<ul>
<li><strong>Scope.</strong> The incident affected &ldquo;fewer than 1% of Notion workspaces.&rdquo;</li>
<li><strong>What Notion used Fireworks for.</strong> Three things: Insights (classification jobs that help customers understand their own AI usage), general inference for serving open-weight models inside Notion, and LEAP, an early-access program whose volunteers&rsquo; data Notion used to experiment with training task-specific models on Fireworks.</li>
<li><strong>What data.</strong> Chat transcripts: any model input or output written to a Notion AI chat.</li>
<li><strong>What was not affected.</strong> AI Meeting Notes audio, transcripts and summaries, unless they were used as input in a Notion AI chat. Notion says metadata it added for analysis (record, workspace, user and thread IDs, dates) was not exposed, and that for most workspaces it added no metadata tying a transcript to a specific workspace, user or company.</li>
<li><strong>Retention.</strong> Notion says its zero data retention commitment did not cover data it had to store to provide features such as chat history and Insights, and adds: &ldquo;Unfortunately, during our investigation we learned that data was retained from the Fireworks endpoints that we used.&rdquo;</li>
<li><strong>What Notion did.</strong> As of November 3 it disconnected its systems from Fireworks, deleted all data in the Fireworks system, stopped sending data and traffic there, halted training runs, disabled Insights during the investigation, stopped serving open-weight models through Fireworks, contacted affected customers and began auditing critical vendors for zero data retention compliance.</li>
</ul>
<h2 id="what-fireworks-has-confirmed">What Fireworks has confirmed</h2>
<p>Fireworks published an <a href="https://fireworks.ai/security/bulletins/october-2026-security-incident-update">October 2026 security incident update</a>, last updated November 9, 2026, on top of its <a href="https://fireworks.ai/security/bulletins/october-2026-security-incident">original October bulletin</a>.</p>
<ul>
<li><strong>First incident.</strong> &ldquo;On October 6, Fireworks identified a security incident involving unauthorized use of internal Fireworks credentials.&rdquo; An unauthorized party accessed environment variables used in evaluation jobs in a limited number of customer accounts, and Fireworks believes API keys and tokens stored as secrets in some of those accounts were captured. Fireworks has not said how the credentials were obtained.</li>
<li><strong>Second path.</strong> &ldquo;A third-party cloud provider that hosted Fireworks R&amp;D machines experienced a security incident on October 23.&rdquo; On October 24 the unauthorized party used that access to reach Fireworks development machines and ran an unauthorized workload that affected customer data; Fireworks says it detected this the same day. It does not name the provider or say how the provider was compromised.</li>
<li><strong>Data accessed.</strong> Resource metadata (identifiers, display names, descriptions, creation times); billing information (addresses, invoice and payment history, balances, credit usage; Fireworks says payment card numbers were not stored and not affected); &ldquo;for certain customers, a subset of incomplete model weights and datasets&rdquo;; and secrets for a limited set of accounts.</li>
<li><strong>Zero data retention products.</strong> &ldquo;We have not detected any impact to Inference, Training SDK and Nexus,&rdquo; which Fireworks describes as zero data retention by default. It says the investigation is ongoing.</li>
<li><strong>Who.</strong> Fireworks links the activity to &ldquo;a cybercriminal threat actor group&rdquo; it describes as financially motivated and focused on selling tokens from unauthorized LLM access to illicit routing gateways, which it calls LLMJacking. It does not name the group.</li>
<li><strong>Response.</strong> Fireworks says it worked with Palo Alto Networks Unit 42, notified law enforcement, revoked and replaced credentials and service accounts, isolated affected systems, tightened access to storage and analytics systems, and disabled self-serve dataset uploads and downloads for Batch Inference and dataset-based training.</li>
</ul>
<h2 id="timeline">Timeline</h2>
<table>
  <thead>
      <tr>
          <th>Date (2026)</th>
          <th>Event</th>
          <th>Source</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Oct 6</td>
          <td>Fireworks identifies unauthorized use of internal credentials; evaluation-job environment variables accessed</td>
          <td>Fireworks bulletin</td>
      </tr>
      <tr>
          <td>Oct 9</td>
          <td>Fireworks publishes its first security bulletin</td>
          <td>Fireworks update page</td>
      </tr>
      <tr>
          <td>Oct 23</td>
          <td>Security incident at a third-party cloud provider hosting Fireworks R&amp;D machines</td>
          <td>Fireworks update</td>
      </tr>
      <tr>
          <td>Oct 24</td>
          <td>Unauthorized workload run on Fireworks development machines; detected the same day</td>
          <td>Fireworks update</td>
      </tr>
      <tr>
          <td>Nov 2</td>
          <td>Fireworks told Notion that Notion AI data had been accessed</td>
          <td>SecurityWeek, citing the companies</td>
      </tr>
      <tr>
          <td>Nov 3</td>
          <td>Notion disconnects its systems from Fireworks</td>
          <td>Notion notice</td>
      </tr>
      <tr>
          <td>Nov 9</td>
          <td>Fireworks updates its bulletin with billing, dataset, model-weight and secrets impact</td>
          <td>Fireworks update</td>
      </tr>
      <tr>
          <td>Nov 11-12</td>
          <td>Disclosure reported by VentureBeat and SecurityWeek</td>
          <td>Press</td>
      </tr>
  </tbody>
</table>
<p>VentureBeat separately reports that Fireworks confirmed a &ldquo;new dataset&rdquo; to Notion on November 8; we could not match that date to either company&rsquo;s own pages, so treat it as the outlet&rsquo;s reporting.</p>
<h2 id="confirmed-vs-not-confirmed">Confirmed vs not confirmed</h2>
<table>
  <thead>
      <tr>
          <th>Claim</th>
          <th>Status</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Fewer than 1% of Notion workspaces affected</td>
          <td>Confirmed by Notion</td>
      </tr>
      <tr>
          <td>Notion AI chat transcripts (inputs and outputs) exposed</td>
          <td>Confirmed by Notion</td>
      </tr>
      <tr>
          <td>Entry via a third-party cloud provider hosting Fireworks R&amp;D machines</td>
          <td>Confirmed by Fireworks</td>
      </tr>
      <tr>
          <td>Fireworks billing data, some datasets and incomplete model weights accessed</td>
          <td>Confirmed by Fireworks</td>
      </tr>
      <tr>
          <td>Fireworks Inference, Training SDK and Nexus unaffected</td>
          <td>Fireworks says no impact detected; investigation ongoing</td>
      </tr>
      <tr>
          <td>Which Fireworks feature stored Notion&rsquo;s transcripts</td>
          <td>Not stated by either company</td>
      </tr>
      <tr>
          <td>Number of Notion customers notified</td>
          <td>Not stated (SecurityWeek says Notion did not give a number)</td>
      </tr>
      <tr>
          <td>Identity of the attacker group</td>
          <td>Not named; Fireworks describes it as a financially motivated LLMJacking group</td>
      </tr>
      <tr>
          <td>Whether the October 6 credential incident and the October 23-24 activity are linked</td>
          <td>Not stated</td>
      </tr>
  </tbody>
</table>
<h2 id="what-this-means-for-engineering-teams">What this means for engineering teams</h2>
<p>Most product teams that ship AI features send prompts to at least one inference vendor. This incident shows where the assumptions break.</p>
<p><strong>Zero data retention is a per-feature property, not a vendor property.</strong> Notion says its own commitment did not cover data it had to store for chat history and Insights, and that it found data retained at the Fireworks endpoints it used. Fireworks says its zero data retention default covers Inference, Training SDK and Nexus. VentureBeat quotes the Fireworks policy page: &ldquo;A few features do store customer content, such as stored responses, batch jobs, and training.&rdquo; Ask each vendor, per endpoint you call (synchronous inference, batch jobs, fine-tuning datasets, evaluations, stored responses), whether content is written to storage, for how long, and how you verify the setting is on. Then check your own account configuration rather than the vendor&rsquo;s default.</p>
<p><strong>Classification and analytics pipelines are data stores too.</strong> The affected Notion feature was Insights, a classification job over chat transcripts, not the chat itself. Batch analytics over prompts tends to create copies that outlive the conversation. Put those pipelines in your data inventory with an owner and a retention period, and include them in <a href="/posts/vetting-third-party-saas-tools-before-they-touch-your-data/">the same vendor review you run for SaaS tools</a>.</p>
<p><strong>Assume users paste secrets into AI chat.</strong> Notion&rsquo;s advice to rotate any credentials pasted into Notion AI is the right default for any chat surface. Secret scanning on prompt logs, and on anything you send to a vendor, catches some of it; our guide to <a href="/posts/how-to-detect-and-rotate-leaked-api-keys-before-attackers-use-them/">detecting and rotating leaked API keys</a> covers the rotation side. The broader patterns are in <a href="/posts/llm-data-leakage-how-ai-applications-accidentally-expose-sensitive-information/">LLM data leakage</a>.</p>
<p><strong>Rotate at the issuer.</strong> Fireworks&rsquo; first bulletin makes a point that applies to every platform with a secrets store: &ldquo;Removing a stored secret alone does not invalidate the underlying credential.&rdquo; If you stored third-party API keys in a vendor&rsquo;s evaluation or job environment, revoke them with the provider that issued them, then check that provider&rsquo;s access logs.</p>
<p><strong>Vendor-of-vendor exposure.</strong> The intrusion path ran through a cloud provider Fireworks used for R&amp;D, two hops from Notion&rsquo;s customers. Your vendor questionnaire should ask whether development and research environments can reach production customer data, because here they apparently could.</p>
<h2 id="limits-of-this-article">Limits of this article</h2>
<p>We relied on the companies&rsquo; own pages and two news outlets. Neither company has named the cloud provider, the attacker group or the storage feature involved, and both say investigations are ongoing. If either company changes its account, we will add a dated update at the top of this post.</p>
<h2 id="sources-and-verification">Sources and verification</h2>
<table>
  <thead>
      <tr>
          <th>Important claim</th>
          <th>Source</th>
          <th>Status</th>
      </tr>
  </thead>
  <tbody>
      <tr>
          <td>Fewer than 1% of workspaces; chat transcripts; what Fireworks was used for; not-affected data; retention statement; actions as of Nov 3; customer guidance</td>
          <td><a href="https://notion.com/help/security-incident-involving-a-third-party-ai-vendor">Notion help center notice</a></td>
          <td>Verified (read Nov 13, 2026; page undated)</td>
      </tr>
      <tr>
          <td>Oct 23 provider incident, Oct 24 workload, data accessed, ZDR products unaffected, LLMJacking attribution, response steps</td>
          <td><a href="https://fireworks.ai/security/bulletins/october-2026-security-incident-update">Fireworks October 2026 security incident update</a>, last updated Nov 9, 2026</td>
          <td>Verified</td>
      </tr>
      <tr>
          <td>Oct 6 internal-credential incident, evaluation-job secrets, &ldquo;Removing a stored secret alone does not invalidate the underlying credential&rdquo;</td>
          <td><a href="https://fireworks.ai/security/bulletins/october-2026-security-incident">Fireworks October 2026 security incident bulletin</a></td>
          <td>Verified</td>
      </tr>
      <tr>
          <td>Fireworks told Notion on Nov 2; Notion did not give a number of affected customers</td>
          <td><a href="https://www.securityweek.com/notion-customer-data-compromised-in-fireworks-ai-hack/">SecurityWeek, Nov 12, 2026</a></td>
          <td>Qualified: press report</td>
      </tr>
      <tr>
          <td>Nov 8 &ldquo;new dataset&rdquo; confirmation; Fireworks policy-page quote on features that store content</td>
          <td><a href="https://venturebeat.com/security/attackers-breached-the-fireworks-ai-storage-where-notion-ai-chats-sat-outside-zero-data-retention">VentureBeat, Nov 11, 2026</a></td>
          <td>Qualified: press report</td>
      </tr>
  </tbody>
</table>
]]></content:encoded></item></channel></rss>