{"slug":"ai-adapt-mq-20261122","entry":{"slug":"ai-adapt-mq-20261122","title":"AI Adaptation Research Report — MQ (Marqeta, Inc.) · BlackSwan constituent","symbol":"MQ","kind":"ai_adaptation","name":"Marqeta, Inc.","book":"AMF-BLACKSWAN","direction":"short","verdict":"blackswan","run_date":"2026-11-22","narrative":"model","level":0,"aas":9.4,"date":"2026-11-22","generated_at":"2026-11-22T07:08:38+00:00","description":"Verdict: BlackSwan (short book). On the 2026-11-22 scoring run, MQ is classed as failing to adapt to AI because the anchor filing contains zero AI sentences, giving it an adoption level of \"absent\".","algorithm_release":"1.1.0","rank":7,"composite":34.883,"story_score":50.06,"market":70.62,"weight":0.085463,"pinned":false},"url":"/marr/ai-adapt-mq-20261122","markdown":"# AI Adaptation Research Report — MQ (Marqeta, Inc.) · BlackSwan constituent\n\n*EventTrader research desk · generated 2026-11-22 07:08 UTC · report ai-adapt-mq-20261122 · narrative: AI analyst, checked against the evidence pack*\n\n| | |\n|---|---|\n| **Company** | MQ — Marqeta, Inc. · SIC 7372 Services-Prepackaged Software |\n| **Book** | AMF-BLACKSWAN (short) — BlackSwan |\n| **Composition algorithm** | Swan Composite v1.1.0 — 50% AI Adaptation Score + 25% AI Story Score + 25% market verdict, conviction-weighted |\n| **Anchor filing** | 10-Q filed 2026-11-03 (period 2026-09-30) |\n| **Adoption level** | 0 — absent · deterministic |\n| **AI Adaptation Score** | 9.4 / 100 · run 2026-11-22 · sector software percentile 1.4 |\n| **Price** | 16.43 USD |\n| **Place in the book** | rank 7 of 10 · composite 34.9 · story 50.1 · market verdict 70.6 · weight 8.5% |\n\n## 1. Verdict\n\n- **Verdict: BlackSwan (short book).** On the 2026-11-22 scoring run, MQ is classed as failing to adapt to AI because the anchor filing contains zero AI sentences, giving it an adoption level of \"absent\" [S1][S5].\n- **Anchor filing:** Marqeta's Form 10-Q filed 2026-11-03 for the quarter ended 2026-09-30 (accession 0001522540-26-000077) [S1].\n- **Adoption level: \"absent\" (level 0)** — the lowest band, driven by a lexical score of 0 and no language-model read [S5].\n- **AI Adaptation Score: 9.428**, placing MQ at the **1.4th percentile among 590 software peers** and the **13.8th percentile across the 6,561-name universe** [S5].\n- **What would move it:** a later filing that actually discusses AI — deployed systems, products, investment, or AI-specific risk language — enough to lift the lexical read and trigger a model review [S1][S5].\n\n## 2. The company\n\nMQ is Marqeta, Inc., SEC filer CIK 1522540, listed on Nasdaq [S4]. The pack classifies it under SIC code 7372, \"Services-Prepackaged Software,\" and assigns it to the software sector; the scoring universe is \"software companies (SIC 7370–7379)\" [S4][S5]. The filing reviewed for this run is Marqeta's quarterly report on Form 10-Q, filed 2026-11-03 and covering the period ending 2026-09-30 [S1]. Beyond filer identity, sector, and this filing, the pack states no further descriptive detail about the business, and this report does not add any. The most recent quote in the pack is $16.43, flagged as not stale [S6].\n\n## 3. What the filing says about AI\n\nThe evidence pack contains **no AI sentence from the anchor filing** — the filing reviewed yielded zero AI sentences (`ai_sentences: 0`), and the pack supplies no grounded quotes to reproduce [S1][S5]. There is therefore nothing to quote verbatim in this section. The filing's substantive text runs to 7,676 words out of 8,773 total, and within that body the score found no substantive AI hits and no AI risk hits [S5]. What the pack cannot tell us is whether AI is discussed anywhere in Marqeta's operations outside this particular 10-Q; the finding is specific to the document scored.\n\n## 4. The AI Adaptation Score\n\nThe adoption level is **\"absent\" (final level 0)**, with a lexical component of 0 [S5]. There is **no model read**: the pack records `model: null` and `model_status: \"none\"`, so there are no sub-scores, no confidence, and no credibility figure to report [S5]. The AI Adaptation Score (AAS) is **9.428**, with a margin of 23.759 and the ticker assigned to the \"non_adopter\" cluster [S5]. Its components, as given, are: level 0.0, greenfield 0.0, credibility 0.0, trajectory 5.958, and investment 3.47 — meaning the modest score comes entirely from the trajectory and investment terms, not from any detected AI adoption, greenfield building, or credibility [S5]. On ranking, MQ sits at the **13.8th percentile of 6,561 names** in the universe and the **1.4th percentile of 590 software-sector peers** [S5]. The sector percentile is the sharper signal here: against its own peer group, MQ scores near the very bottom.\n\n## 5. Trajectory\n\nThe previous periodic report was Marqeta's Form 10-Q filed 2026-08-04 (accession 0001522540-26-000063), and it was also scored at **level 0, \"absent,\" with zero AI sentences** and substantive density of 0.0 [S3]. The adoption level therefore **did not move** between the two quarters — \"absent\" in the prior filing and \"absent\" in the anchor filing [S1][S3]. The score's trajectory component (5.958) is a model input reported as given and should not be read as a change in adoption level, which was flat [S5].\n\n## 6. Recent developments\n\n- **2026-10-12** — \"Pending Class Action Lawsuit Involving the Purchase, Acquisition and/or Sale of Marqeta Inc. Class A Common Stock, Call Options and/or Put Options Between February 28, 2024 and November 4, 2024\" (wire:prnewswire) [S5].\n\nThis is the only headline in the pack's 45-day window [S5]. It concerns securities litigation, not AI adoption, and does not bear on the score.\n\n## 7. What would change the verdict\n\nTo move MQ out of the BlackSwan book, a later filing or model read would need to show AI adoption the score can actually see. Concretely, within the four corners of a future periodic filing: substantive AI sentences (the anchor filing had zero), AI language placed in substance sections rather than boilerplate, and ideally evidence of deployed or proprietary AI systems, AI-specific investment, or AI risk disclosure — each of which the current filing registers at zero [S1][S5]. Because the level is \"absent\" and no language model has yet read the document (`model_status: \"none\"`), a filing with enough AI content to clear the lexical threshold could also trigger a model review, which is the path to a credibility and sub-score profile the current run lacks [S5]. None of this is a forecast; it is what the components would have to register to change.\n\n## 8. Risks and caveats\n\nA filing-based score sees only what a company writes in the specific document scored. It **can** count AI language, locate it, and weigh it; it **cannot** see AI work that Marqeta does but does not disclose in this 10-Q, nor AI embedded in products described in other words. A keyword and model read of one quarterly filing is not a full audit of the company's technology. The verdict would be **falsified** if the anchor filing in fact contains substantive AI discussion the pack failed to capture — but on the evidence given (`ai_sentences: 0`, `lexical: 0`, across 7,676 substantive words), the \"absent\" level is internally consistent [S1][S5]. The headline risk item in the pack is litigation, which the score does not incorporate and which readers should weigh separately [S5].\n\n## 9. Watch list\n\n- **Next filing:** Marqeta's next periodic report after the 2026-09-30 10-Q; the EDGAR submissions feed is the place to catch it [S4].\n- **Model read:** none has run on the anchor filing (`model_status: \"none\"`); a future filing with AI content is what would prompt one [S5].\n- **Next scoring run:** this verdict reflects the 2026-11-22 run and is effective from 2026-11-04; subsequent runs will re-read the latest anchor filing [S1][S5].\n## Place in the book\n\n- Rank 7 of 10 in AMF-BLACKSWAN (Swan Composite v1.1.0, scoring run 2026-11-22): the book holds the lowest composites [S5].\n- Composite: 0.50 × 9.4 (AAS) + 0.25 × 50.1 (story) + 0.25 × 70.6 (market) = 34.9 [S5].\n- Market verdict 70.6: over the last 90 days MQ rose more and fell less than the software universe's median daily return (down-capture 0.95, up-capture 1.77, L = 0.82; 36 down / 18 up days) — the Failure-To-Adapt fund's selection signal, 50 = moves with the software universe's median daily return [S5].\n- AI Story Score 50.1: no AI story or a neutral one in the AI stories written about the company [S5].\n- Weight: rank factor 1.33 (rank 7 of 10) × story factor 1.00 × market factor 1.00, normalised across the book and clamped to 5–20% → 8.5% [S5].\n\n## Sources\n\n- **S1** — MQ 10-Q filed 2026-11-03 — SEC primary document (the anchor filing the score reads) — <https://www.sec.gov/Archives/edgar/data/1522540/000152254026000077/mq-20260930.htm>\n- **S2** — EDGAR filing index for accession 0001522540-26-000077 — <https://www.sec.gov/Archives/edgar/data/1522540/000152254026000077/>\n- **S3** — MQ 10-Q filed 2026-08-04 — the previous periodic report (trajectory) — <https://www.sec.gov/Archives/edgar/data/1522540/000152254026000063/>\n- **S4** — SEC EDGAR company submissions — Marqeta, Inc. (CIK 1522540) — <https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=1522540>\n- **S5** — EventTrader AI Adaptation Score, scoring run 2026-11-22 — the AMF-BLACKSWAN fund page (method, composition) — <https://cymetica.com/fund-performance/AMF-BLACKSWAN>\n- **S6** — EventTrader real-price feed — MQ quote — <https://cymetica.com/api/v1/prices/MQ>\n\n## Method and disclaimer\n\nThe AI Adaptation Score reads each company's own periodic SEC reports (10-K, 10-Q, 20-F, 40-F): deterministic features (AI sentences by section, proprietary-model, deployment, outcome, investment and vendor language, boilerplate and negation excluded) set an adoption level on a 0–4 rubric; a model read of the filing's AI sentences confirms the level and scores credibility and greenfield (the business built with AI from the ground up rather than retrofitted); the score combines level, greenfield, credibility, trajectory against the previous report and investment, measured against the company's sector, with greenfield carrying the highest reward. The books are composed by Swan Composite v1.1.0: each company's composite is 50% AI Adaptation Score + 25% AI Story Score + 25% market verdict — the AI Story Score is the sentiment of the AI stories written about it, the belief network of companies named in the same stories and the price reaction to those stories (no story scores neutral); the market verdict is the Failure-To-Adapt fund's selection signal, the signed down/up capture ratio of the stock against the software universe's median return over 90 days (a stock that falls more and rises less than its peers scores below 50). AMF-BLACKSWAN holds the ten lowest composites that pass the liquidity screen and the new-listing gate and AMF-WHITESWAN the ten highest, weighted by conviction (rank, story agreement and market agreement, each leg 5–20%), recomposed on every scoring run. Every quote in this report is the filing's text verbatim; every factual sentence names its source; a source the platform could not reach is reported as missing, never filled in.\n\n*This report is information, not investment advice. Figures come from the cited sources as of their dates. Nothing here is a recommendation to buy, sell or hold any asset.*\n","html":"<h1>AI Adaptation Research Report — MQ (Marqeta, Inc.) · BlackSwan constituent</h1>\n<p><em>EventTrader research desk · generated 2026-11-22 07:08 UTC · report ai-adapt-mq-20261122 · narrative: AI analyst, checked against the evidence pack</em></p>\n<table>\n<thead>\n<tr>\n<th></th>\n<th></th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td><strong>Company</strong></td>\n<td>MQ — Marqeta, Inc. · SIC 7372 Services-Prepackaged Software</td>\n</tr>\n<tr>\n<td><strong>Book</strong></td>\n<td>AMF-BLACKSWAN (short) — BlackSwan</td>\n</tr>\n<tr>\n<td><strong>Composition algorithm</strong></td>\n<td>Swan Composite v1.1.0 — 50% AI Adaptation Score + 25% AI Story Score + 25% market verdict, conviction-weighted</td>\n</tr>\n<tr>\n<td><strong>Anchor filing</strong></td>\n<td>10-Q filed 2026-11-03 (period 2026-09-30)</td>\n</tr>\n<tr>\n<td><strong>Adoption level</strong></td>\n<td>0 — absent · deterministic</td>\n</tr>\n<tr>\n<td><strong>AI Adaptation Score</strong></td>\n<td>9.4 / 100 · run 2026-11-22 · sector software percentile 1.4</td>\n</tr>\n<tr>\n<td><strong>Price</strong></td>\n<td>16.43 USD</td>\n</tr>\n<tr>\n<td><strong>Place in the book</strong></td>\n<td>rank 7 of 10 · composite 34.9 · story 50.1 · market verdict 70.6 · weight 8.5%</td>\n</tr>\n</tbody>\n</table>\n<h2>1. Verdict</h2>\n<ul>\n<li><strong>Verdict: BlackSwan (short book).</strong> On the 2026-11-22 scoring run, MQ is classed as failing to adapt to AI because the anchor filing contains zero AI sentences, giving it an adoption level of \"absent\" [S1][S5].</li>\n<li><strong>Anchor filing:</strong> Marqeta's Form 10-Q filed 2026-11-03 for the quarter ended 2026-09-30 (accession 0001522540-26-000077) [S1].</li>\n<li><strong>Adoption level: \"absent\" (level 0)</strong> — the lowest band, driven by a lexical score of 0 and no language-model read [S5].</li>\n<li><strong>AI Adaptation Score: 9.428</strong>, placing MQ at the <strong>1.4th percentile among 590 software peers</strong> and the <strong>13.8th percentile across the 6,561-name universe</strong> [S5].</li>\n<li><strong>What would move it:</strong> a later filing that actually discusses AI — deployed systems, products, investment, or AI-specific risk language — enough to lift the lexical read and trigger a model review [S1][S5].</li>\n</ul>\n<h2>2. The company</h2>\n<p>MQ is Marqeta, Inc., SEC filer CIK 1522540, listed on Nasdaq [S4]. The pack classifies it under SIC code 7372, \"Services-Prepackaged Software,\" and assigns it to the software sector; the scoring universe is \"software companies (SIC 7370–7379)\" [S4][S5]. The filing reviewed for this run is Marqeta's quarterly report on Form 10-Q, filed 2026-11-03 and covering the period ending 2026-09-30 [S1]. Beyond filer identity, sector, and this filing, the pack states no further descriptive detail about the business, and this report does not add any. The most recent quote in the pack is $16.43, flagged as not stale [S6].</p>\n<h2>3. What the filing says about AI</h2>\n<p>The evidence pack contains <strong>no AI sentence from the anchor filing</strong> — the filing reviewed yielded zero AI sentences (<code>ai_sentences: 0</code>), and the pack supplies no grounded quotes to reproduce [S1][S5]. There is therefore nothing to quote verbatim in this section. The filing's substantive text runs to 7,676 words out of 8,773 total, and within that body the score found no substantive AI hits and no AI risk hits [S5]. What the pack cannot tell us is whether AI is discussed anywhere in Marqeta's operations outside this particular 10-Q; the finding is specific to the document scored.</p>\n<h2>4. The AI Adaptation Score</h2>\n<p>The adoption level is <strong>\"absent\" (final level 0)</strong>, with a lexical component of 0 [S5]. There is <strong>no model read</strong>: the pack records <code>model: null</code> and <code>model_status: \"none\"</code>, so there are no sub-scores, no confidence, and no credibility figure to report [S5]. The AI Adaptation Score (AAS) is <strong>9.428</strong>, with a margin of 23.759 and the ticker assigned to the \"non_adopter\" cluster [S5]. Its components, as given, are: level 0.0, greenfield 0.0, credibility 0.0, trajectory 5.958, and investment 3.47 — meaning the modest score comes entirely from the trajectory and investment terms, not from any detected AI adoption, greenfield building, or credibility [S5]. On ranking, MQ sits at the <strong>13.8th percentile of 6,561 names</strong> in the universe and the <strong>1.4th percentile of 590 software-sector peers</strong> [S5]. The sector percentile is the sharper signal here: against its own peer group, MQ scores near the very bottom.</p>\n<h2>5. Trajectory</h2>\n<p>The previous periodic report was Marqeta's Form 10-Q filed 2026-08-04 (accession 0001522540-26-000063), and it was also scored at <strong>level 0, \"absent,\" with zero AI sentences</strong> and substantive density of 0.0 [S3]. The adoption level therefore <strong>did not move</strong> between the two quarters — \"absent\" in the prior filing and \"absent\" in the anchor filing [S1][S3]. The score's trajectory component (5.958) is a model input reported as given and should not be read as a change in adoption level, which was flat [S5].</p>\n<h2>6. Recent developments</h2>\n<ul>\n<li><strong>2026-10-12</strong> — \"Pending Class Action Lawsuit Involving the Purchase, Acquisition and/or Sale of Marqeta Inc. Class A Common Stock, Call Options and/or Put Options Between February 28, 2024 and November 4, 2024\" (wire:prnewswire) [S5].</li>\n</ul>\n<p>This is the only headline in the pack's 45-day window [S5]. It concerns securities litigation, not AI adoption, and does not bear on the score.</p>\n<h2>7. What would change the verdict</h2>\n<p>To move MQ out of the BlackSwan book, a later filing or model read would need to show AI adoption the score can actually see. Concretely, within the four corners of a future periodic filing: substantive AI sentences (the anchor filing had zero), AI language placed in substance sections rather than boilerplate, and ideally evidence of deployed or proprietary AI systems, AI-specific investment, or AI risk disclosure — each of which the current filing registers at zero [S1][S5]. Because the level is \"absent\" and no language model has yet read the document (<code>model_status: \"none\"</code>), a filing with enough AI content to clear the lexical threshold could also trigger a model review, which is the path to a credibility and sub-score profile the current run lacks [S5]. None of this is a forecast; it is what the components would have to register to change.</p>\n<h2>8. Risks and caveats</h2>\n<p>A filing-based score sees only what a company writes in the specific document scored. It <strong>can</strong> count AI language, locate it, and weigh it; it <strong>cannot</strong> see AI work that Marqeta does but does not disclose in this 10-Q, nor AI embedded in products described in other words. A keyword and model read of one quarterly filing is not a full audit of the company's technology. The verdict would be <strong>falsified</strong> if the anchor filing in fact contains substantive AI discussion the pack failed to capture — but on the evidence given (<code>ai_sentences: 0</code>, <code>lexical: 0</code>, across 7,676 substantive words), the \"absent\" level is internally consistent [S1][S5]. The headline risk item in the pack is litigation, which the score does not incorporate and which readers should weigh separately [S5].</p>\n<h2>9. Watch list</h2>\n<ul>\n<li><strong>Next filing:</strong> Marqeta's next periodic report after the 2026-09-30 10-Q; the EDGAR submissions feed is the place to catch it [S4].</li>\n<li><strong>Model read:</strong> none has run on the anchor filing (<code>model_status: \"none\"</code>); a future filing with AI content is what would prompt one [S5].</li>\n<li><strong>Next scoring run:</strong> this verdict reflects the 2026-11-22 run and is effective from 2026-11-04; subsequent runs will re-read the latest anchor filing [S1][S5].</li>\n</ul>\n<h2>Place in the book</h2>\n<ul>\n<li>Rank 7 of 10 in AMF-BLACKSWAN (Swan Composite v1.1.0, scoring run 2026-11-22): the book holds the lowest composites [S5].</li>\n<li>Composite: 0.50 × 9.4 (AAS) + 0.25 × 50.1 (story) + 0.25 × 70.6 (market) = 34.9 [S5].</li>\n<li>Market verdict 70.6: over the last 90 days MQ rose more and fell less than the software universe's median daily return (down-capture 0.95, up-capture 1.77, L = 0.82; 36 down / 18 up days) — the Failure-To-Adapt fund's selection signal, 50 = moves with the software universe's median daily return [S5].</li>\n<li>AI Story Score 50.1: no AI story or a neutral one in the AI stories written about the company [S5].</li>\n<li>Weight: rank factor 1.33 (rank 7 of 10) × story factor 1.00 × market factor 1.00, normalised across the book and clamped to 5–20% → 8.5% [S5].</li>\n</ul>\n<h2>Sources</h2>\n<ul>\n<li><strong>S1</strong> — MQ 10-Q filed 2026-11-03 — SEC primary document (the anchor filing the score reads) — <a href=\"https://www.sec.gov/Archives/edgar/data/1522540/000152254026000077/mq-20260930.htm\" rel=\"noopener noreferrer\">https://www.sec.gov/Archives/edgar/data/1522540/000152254026000077/mq-20260930.htm</a></li>\n<li><strong>S2</strong> — EDGAR filing index for accession 0001522540-26-000077 — <a href=\"https://www.sec.gov/Archives/edgar/data/1522540/000152254026000077/\" rel=\"noopener noreferrer\">https://www.sec.gov/Archives/edgar/data/1522540/000152254026000077/</a></li>\n<li><strong>S3</strong> — MQ 10-Q filed 2026-08-04 — the previous periodic report (trajectory) — <a href=\"https://www.sec.gov/Archives/edgar/data/1522540/000152254026000063/\" rel=\"noopener noreferrer\">https://www.sec.gov/Archives/edgar/data/1522540/000152254026000063/</a></li>\n<li><strong>S4</strong> — SEC EDGAR company submissions — Marqeta, Inc. (CIK 1522540) — <a href=\"https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&amp;CIK=1522540\" rel=\"noopener noreferrer\">https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&amp;CIK=1522540</a></li>\n<li><strong>S5</strong> — EventTrader AI Adaptation Score, scoring run 2026-11-22 — the AMF-BLACKSWAN fund page (method, composition) — <a href=\"https://cymetica.com/fund-performance/AMF-BLACKSWAN\" rel=\"noopener noreferrer\">https://cymetica.com/fund-performance/AMF-BLACKSWAN</a></li>\n<li><strong>S6</strong> — EventTrader real-price feed — MQ quote — <a href=\"https://cymetica.com/api/v1/prices/MQ\" rel=\"noopener noreferrer\">https://cymetica.com/api/v1/prices/MQ</a></li>\n</ul>\n<h2>Method and disclaimer</h2>\n<p>The AI Adaptation Score reads each company's own periodic SEC reports (10-K, 10-Q, 20-F, 40-F): deterministic features (AI sentences by section, proprietary-model, deployment, outcome, investment and vendor language, boilerplate and negation excluded) set an adoption level on a 0–4 rubric; a model read of the filing's AI sentences confirms the level and scores credibility and greenfield (the business built with AI from the ground up rather than retrofitted); the score combines level, greenfield, credibility, trajectory against the previous report and investment, measured against the company's sector, with greenfield carrying the highest reward. The books are composed by Swan Composite v1.1.0: each company's composite is 50% AI Adaptation Score + 25% AI Story Score + 25% market verdict — the AI Story Score is the sentiment of the AI stories written about it, the belief network of companies named in the same stories and the price reaction to those stories (no story scores neutral); the market verdict is the Failure-To-Adapt fund's selection signal, the signed down/up capture ratio of the stock against the software universe's median return over 90 days (a stock that falls more and rises less than its peers scores below 50). AMF-BLACKSWAN holds the ten lowest composites that pass the liquidity screen and the new-listing gate and AMF-WHITESWAN the ten highest, weighted by conviction (rank, story agreement and market agreement, each leg 5–20%), recomposed on every scoring run. Every quote in this report is the filing's text verbatim; every factual sentence names its source; a source the platform could not reach is reported as missing, never filled in.</p>\n<p><em>This report is information, not investment advice. Figures come from the cited sources as of their dates. Nothing here is a recommendation to buy, sell or hold any asset.</em></p>\n"}