{"compiledCode":"\"use strict\";Object.defineProperty(exports, \"__esModule\", {value: true}); function _interopRequireDefault(obj) { return obj && obj.__esModule ? obj : { default: obj }; }var _react = require('react'); var _react2 = _interopRequireDefault(_react);\n\n function RiskPage() {\n  const d = {\"h1\":\"Risk and brand safety in AI answers\",\"intro\":\"What to do when an assistant says something false about your brand, why the manipulation tactics spreading through this category are liabilities rather than tactics, and how to protect prompt sets and client data while you measure.\",\"notice\":\"This page is general information, not legal advice; for anything jurisdiction-specific - defamation, advertising rules, disclosure obligations, or data protection - consult qualified counsel in the relevant jurisdiction.\",\"primaryCtas\":[{\"label\":\"Read the correction playbook\",\"href\":\"/blog/false-claim-playbook\"},{\"label\":\"Use the correction protocol\",\"href\":\"/corrections\"}],\"inShort\":\"This reference covers four things: an incident sequence for false brand statements, from repeat runs through source correction and re-measurement; why manipulation tactics create detection, enforcement, regulatory, and reputational risk; prompt injection as a defensive threat model; and data hygiene questions for measurement vendors. It closes with a method for maintaining a dated policy and regulation record. General information, not legal advice.\",\"whoFor\":[\"Brand and reputation managers who have found an assistant stating something untrue about their company\",\"Founders and heads of marketing weighing whether a competitor is gaining an unfair advantage and what to do about it\",\"SEO and growth leads being pitched tactics that sound clever and feel wrong\",\"Analytics, security, and legal partners assessing what happens to prompt sets and client data during measurement\",\"Agency leads who need a defensible position to give clients asking for shortcuts\"],\"whenToUse\":[\"An assistant has described your company inaccurately and someone senior has seen it\",\"A vendor or contractor has proposed a tactic aimed at influencing what assistants say\",\"You are drafting a measurement vendor's security and data-handling review\",\"You need to explain to a stakeholder why manipulation is a business risk rather than an edge\",\"You are tracking how platform policy and advertising regulation may affect this work\"],\"sections\":[{\"heading\":\"When an assistant states something false about your brand\",\"paragraphs\":[\"The instinct is to screenshot it and escalate. Resist that for an hour, because a single answer is not a measurement, and an escalation built on one screenshot tends to collapse when someone else asks the same question and gets a correct answer.\",\"Work in sequence. Document with repeat runs across several phrasings and more than one assistant, recording dates, model versions, and how often the false statement appears. Establish frequency, because a claim appearing in most runs is a different problem from one appearing occasionally. Trace the likely source: an outdated page of your own, a stale third-party listing, a news item, a merged or confused entity, or a genuine model error with no traceable origin.\",\"Then correct the source rather than the model. You have no control over a model's output, and any vendor implying otherwise is selling something they cannot deliver. What you can do is fix your own pages, request corrections from third parties carrying the wrong information, make your entity description consistent everywhere it appears, and use any provider feedback channel that exists.\",\"Finally, re-measure on a schedule. Change here follows two clocks - retrieval, which can update relatively quickly, and model training, which does not - so set expectations accordingly and record the re-runs so you can show what changed and when.\"],\"bullets\":[{\"text\":\"The full incident sequence, step by step\",\"href\":\"/blog/false-claim-playbook\"},{\"text\":\"Making your entity description consistent across sources\",\"href\":\"/get-recommended\"},{\"text\":\"How many runs it takes to establish frequency\",\"href\":\"/measure\"},{\"text\":\"When a comparison claim becomes a legal question\",\"href\":\"/blog/competitor-claims-legal\"}],\"note\":\"Document before you escalate. An incident file with dated runs, model versions, and a frequency estimate is useful to counsel, to a provider, and to your own executives. A screenshot is not.\"},{\"heading\":\"Why manipulation tactics are liabilities\",\"paragraphs\":[\"A market is forming around techniques meant to influence what assistants say by degrading the record they read: coordinated inauthentic posting, fake or incentivized reviews, content networks built to be cited, and text planted to interfere with how a model handles a page. This reference covers these as risks. It does not explain how to carry them out and contains no operational detail.\",\"The case against them is practical rather than moral, and it holds on four grounds at once.\",\"Detection: platforms, review sites, and providers invest continuously in identifying coordinated inauthentic behaviour, and detection improves faster than evasion. Enforcement: the consequences land on the beneficiary brand, not the contractor who executed the work - removal, delisting, account termination, and loss of a source you may have spent years earning. Regulatory exposure: advertising and consumer-protection regimes in several jurisdictions treat fake or undisclosed endorsements as actionable, and that exposure sits with the advertiser. Reputational cost: the discovery of a campaign is itself a story, and it lands in exactly the sources that decide your category's answers.\",\"There is a quieter cost too. Effort spent on manipulation is effort not spent on the legitimate work - accurate documentation, real third-party coverage, retrievable pages - which continues compounding after the shortcut has been rolled back.\"],\"bullets\":[{\"text\":\"The defensive taxonomy of tactics and their consequences\",\"href\":\"/blog/manipulation-tactics-in-ai-seo\"},{\"text\":\"How review fraud and astroturfing get detected\",\"href\":\"/blog/review-fraud-and-astroturfing\"},{\"text\":\"The legitimate levers, with honest time-to-effect\",\"href\":\"/get-recommended\"}],\"note\":\"If a vendor promises placement in AI recommendations, treat the promise itself as the warning. Nobody controls a model's output, and the methods used to imply otherwise are usually the ones described above.\"},{\"heading\":\"Prompt injection as a threat model for brand owners\",\"paragraphs\":[\"Prompt injection is the general problem of untrusted text reaching a model in a position where it is treated as instruction rather than as content. For a brand owner, the relevant question is not how it works mechanically but where exposure sits. This reference publishes no example payloads or techniques.\",\"Two exposures matter. Inbound: content you host or accept - user comments, reviews, uploaded documents, community posts, third-party widgets - can carry text intended for a machine reader rather than a human one. Outbound: pages about you that you do not control may carry the same, and it may affect how a retriever summarises you.\",\"The defensive posture is unglamorous and familiar. Know what user-generated content you host and how it is moderated. Review what your own pages contain, including hidden or templated text you did not write. Watch for descriptions of your brand in answers that do not correspond to anything in your public record, because an inexplicable summary is worth investigating. Bring your security team into the conversation, since this overlaps with their existing work on untrusted input.\",\"It is also worth saying that providers treat this as an active security problem on their side. Your job is to reduce what you contribute to the attack surface, not to defend the model.\"],\"bullets\":[{\"text\":\"Prompt injection explained defensively, for brand owners\",\"href\":\"/blog/prompt-injection-explained\"},{\"text\":\"Auditing what a retriever can actually see on your site\",\"href\":\"/ai-readiness\"}]},{\"heading\":\"Data hygiene when you measure\",\"paragraphs\":[\"Measuring AI visibility means sending prompts somewhere. Those prompts encode your category framing, your competitive set, and sometimes your clients' names - which makes them commercially sensitive even though they look like harmless questions.\",\"Ask any measurement vendor the same set of questions and get the answers in writing. Treat missing, vague, or contradictory answers as an unresolved procurement risk.\"],\"bullets\":[{\"text\":\"Where are prompt sets and results stored, and under whose control?\"},{\"text\":\"How long is data retained, and can you require deletion?\"},{\"text\":\"Is any customer data used to train models, improve products, or seed benchmarks?\"},{\"text\":\"Can your prompt set, results, or brand appear in the vendor's public research or marketing, and on what permission basis?\"},{\"text\":\"Which sub-processors and model providers see the data, and under what terms?\"},{\"text\":\"How is collection performed - official provider APIs, or something that could stop working or breach a provider's terms?\",\"href\":\"/blog/measurement-data-hygiene\"},{\"text\":\"RankRift's own site privacy policy\",\"href\":\"/privacy-policy\"}]},{\"heading\":\"Policy and regulation watch\",\"paragraphs\":[\"Two sets of rules move underneath this work. Platform and provider policy governs what is permitted in measurement and what happens to content and accounts caught manipulating recommendations. Public regulation - advertising standards, consumer protection, endorsement and disclosure rules, and emerging AI transparency requirements - governs what brands may claim and how endorsements must be disclosed.\",\"Both change and vary by jurisdiction. Maintain a dated record that distinguishes proposals from rules in force and links directly to primary documents. Record jurisdiction, issuing authority, document date, effective date, status, source URL, and last-check date.\",\"Nothing here is a substitute for advice on your own situation. Where a decision turns on defamation, advertising claims, takedown routes, endorsement disclosure, or data protection, get qualified counsel in the relevant jurisdiction before acting.\"],\"bullets\":[{\"text\":\"A monitoring protocol for policy and regulation changes\",\"href\":\"/blog/regulation-watch\"},{\"text\":\"Editorial and sourcing specification\",\"href\":\"/editorial-policy\"},{\"text\":\"Correction and re-run record\",\"href\":\"/corrections\"}],\"note\":\"General information only, and not legal advice.\"},{\"heading\":\"The rest of this cluster\",\"paragraphs\":[\"Start with the correction playbook if you have a live incident. The rest is best read before you need it.\"],\"bullets\":[{\"text\":\"An AI assistant is wrong about your brand: a correction playbook\",\"href\":\"/blog/false-claim-playbook\"},{\"text\":\"The manipulation tactics spreading in AI SEO, and why each is a liability\",\"href\":\"/blog/manipulation-tactics-in-ai-seo\"},{\"text\":\"Review fraud and astroturfing: the detection and consequence picture\",\"href\":\"/blog/review-fraud-and-astroturfing\"},{\"text\":\"Prompt injection, explained defensively\",\"href\":\"/blog/prompt-injection-explained\"},{\"text\":\"Competitor comparisons in AI answers: when it becomes a legal problem\",\"href\":\"/blog/competitor-claims-legal\"},{\"text\":\"What your data leaves behind when you measure AI visibility\",\"href\":\"/blog/measurement-data-hygiene\"},{\"text\":\"Disclosure and regulation watch\",\"href\":\"/blog/regulation-watch\"}]}],\"related\":[{\"label\":\"Corrections and disputes\",\"href\":\"/corrections\"},{\"label\":\"Editorial policy and independence\",\"href\":\"/editorial-policy\"},{\"label\":\"The legitimate levers\",\"href\":\"/get-recommended\"},{\"label\":\"Measurement and metric definitions\",\"href\":\"/measure\"},{\"label\":\"Reporting and causal limits\",\"href\":\"/reporting\"},{\"label\":\"Privacy and data handling\",\"href\":\"/privacy-policy\"}],\"faqs\":[{\"q\":\"An assistant said we were acquired and we weren't. What do I do first?\",\"a\":\"Establish frequency before you escalate. Run the question again in several phrasings, across more than one assistant, recording dates and model versions, and note how often the false statement appears. Then look for the origin: an old press item, a stale directory listing, a confused entity with a similar name, or your own outdated page. Correct that source and any third party repeating it, then re-measure on a schedule. Document everything as you go, because that file is what makes any later escalation credible.\"},{\"q\":\"Can I get an assistant to retract something about my company?\",\"a\":\"You cannot compel a model's output, and nobody honestly can. What exists is indirect: providers offer feedback channels of varying usefulness, and correcting the underlying sources changes what a retrieving assistant finds. Where a statement is seriously damaging, this stops being a marketing problem and becomes a legal one, with routes that depend heavily on jurisdiction. That is a conversation for qualified counsel. This page is general information and cannot tell you what applies where you operate.\"},{\"q\":\"Our competitor is obviously manipulating this. Should we respond in kind?\",\"a\":\"No, and the reason is practical rather than moral. Detection improves continuously, enforcement lands on the beneficiary brand rather than the contractor, several jurisdictions treat fake or undisclosed endorsements as actionable against the advertiser, and discovery becomes a story in exactly the sources that decide your category's answers. Meanwhile the effort would have to be spent again after any rollback. Document what you observe, keep your own record accurate, and let the compounding legitimate work do the job.\"},{\"q\":\"Why won't you explain how these tactics work in detail?\",\"a\":\"Because a detailed description is an instruction manual regardless of the framing around it, and this site would become part of the problem it documents. Understanding that a category of tactic exists, how it tends to be detected, and what it costs when it fails is enough for a brand owner to make a decision and to recognise a bad pitch. Anything more operational serves practitioners rather than defenders. That line applies to every post in this cluster.\"},{\"q\":\"Should I worry about prompt injection if we don't build AI products?\",\"a\":\"Mildly, and mostly about your own surface area. If you host user-generated content, accept uploads, run a community, or embed third-party widgets, you host text a machine reader may encounter. Pages about you that you do not control carry the same possibility. The practical response is ordinary hygiene: know what untrusted content you publish, review your own pages for text you did not write, and investigate any description of your brand in an answer that does not match your public record.\"},{\"q\":\"Are my prompts and client names safe with a measurement vendor?\",\"a\":\"Assume nothing and ask in writing. Prompt sets encode your category framing and competitive set, so they are commercially sensitive even though they look like plain questions. Ask where data is stored, how long it is retained, whether deletion can be required, whether anything is used for training or product improvement, whether your brand can appear in the vendor's public research, and which sub-processors and model providers see it. A vendor that cannot answer quickly has not thought about it.\"},{\"q\":\"Is any of this regulated yet?\",\"a\":\"Parts of it, unevenly, and the answer varies by jurisdiction. Advertising and consumer-protection rules can cover endorsements and fake reviews regardless of channel. AI-specific transparency rules may be proposed or in force depending on place and date. Check the relevant authority's primary text, record its status and effective date, and get qualified counsel for what applies to you; this page is general information only.\"}],\"closingCtaHeading\":\"Have the playbook before the incident\",\"closingCtaText\":\"The teams that handle a false model claim well are the ones who already knew to document with repeat runs before escalating. Read the playbook now, not on the day it happens.\",\"closingCtas\":[{\"label\":\"Read the correction playbook\",\"href\":\"/blog/false-claim-playbook\"},{\"label\":\"Use the correction protocol\",\"href\":\"/corrections\"},{\"label\":\"Use the briefing template\",\"href\":\"/newsletter\"}]};\n  const btn = (kind) => ({\n    display: \"inline-block\",\n    padding: kind === \"primary\" ? \"11px 18px\" : \"9px 16px\",\n    margin: \"6px 10px 6px 0\",\n    borderRadius: 8,\n    textDecoration: \"none\",\n    fontWeight: 700,\n    fontSize: 15,\n    background: kind === \"primary\" ? \"#3d3f8c\" : \"transparent\",\n    color: kind === \"primary\" ? \"#ffffff\" : \"#3d3f8c\",\n    border: \"2px solid #3d3f8c\",\n  });\n  return (\n    _react2.default.createElement('div', { className: [\"np-rankrift\",\"rrpage\"].join(\" \"), 'aria-labelledby': \"rr-title\",}\n, _react2.default.createElement('style', null, \"\\n        .np-rankrift.rrpage{max-width:880px;margin:0 auto;padding:28px 20px 64px;color:#1a1f2b;\\n          font-family:var(--cdqcat5-font-body, \\\"DM Sans\\\", Arial, sans-serif);\\n          line-height:1.62;font-size:17px}\\n        .np-rankrift.rrpage h1{font-size:2rem;line-height:1.2;margin:0 0 12px}\\n        .np-rankrift.rrpage h2{font-size:1.35rem;line-height:1.3;margin:36px 0 10px}\\n        .np-rankrift.rrpage p{margin:0 0 14px}\\n        .np-rankrift.rrpage ul{margin:0 0 16px;padding-left:22px}\\n        .np-rankrift.rrpage li{margin:0 0 8px}\\n        .np-rankrift.rrpage a{color:#33368a}\\n        .np-rankrift .rr-intro{font-size:1.16rem;color:#39414f}\\n        .np-rankrift .rr-callout{background:#f2f3fb;border-left:4px solid #3d3f8c;padding:12px 16px;border-radius:6px;margin:6px 0 16px;font-size:0.98rem}\\n        .np-rankrift.rrpage details{border:1px solid #e3e5ef;border-radius:8px;padding:10px 14px;margin:0 0 10px;background:#fbfbfe}\\n        .np-rankrift.rrpage summary{cursor:pointer;font-weight:600}\\n        .np-rankrift.rrpage details p{margin:10px 0 2px}\\n        .np-rankrift .rr-ctarow{margin:16px 0 10px}\\n        @media (max-width:560px){.np-rankrift.rrpage{font-size:16px;padding:22px 16px 56px}.np-rankrift.rrpage h1{font-size:1.6rem}}\\n      \")\n\n      , _react2.default.createElement('h1', { id: \"rr-title\",}, d.h1)\n      , _react2.default.createElement('p', { className: \"rr-intro\",}, d.intro)\n      , _react2.default.createElement('div', { className: \"rr-ctarow\",}\n        , d.primaryCtas.map((c, i) => (\n          _react2.default.createElement('a', { key: i, href: c.href, style: btn(i === 0 ? \"primary\" : \"ghost\"),}, c.label)\n        ))\n      )\n\n      , d.notice ? _react2.default.createElement('div', { className: \"rr-callout\",}, d.notice) : null\n\n      , d.inShort ? (\n        _react2.default.createElement('section', { 'aria-label': \"In short\" ,}\n          , _react2.default.createElement('h2', null, \"In short\" )\n          , _react2.default.createElement('p', null, d.inShort)\n        )\n      ) : null\n\n      , d.whoFor && d.whoFor.length ? (\n        _react2.default.createElement('section', { 'aria-label': \"Who this is for\"   ,}\n          , _react2.default.createElement('h2', null, \"Who this is for\"   )\n          , _react2.default.createElement('ul', null, d.whoFor.map((x, i) => _react2.default.createElement('li', { key: i,}, x)))\n        )\n      ) : null\n\n      , d.whenToUse && d.whenToUse.length ? (\n        _react2.default.createElement('section', { 'aria-label': \"When to use this\"   ,}\n          , _react2.default.createElement('h2', null, \"When to use this\"   )\n          , _react2.default.createElement('ul', null, d.whenToUse.map((x, i) => _react2.default.createElement('li', { key: i,}, x)))\n        )\n      ) : null\n\n      , d.sections.map((s, i) => (\n        _react2.default.createElement('section', { key: i,}\n          , _react2.default.createElement('h2', { id: s.heading.toLowerCase().replace(/[^a-z0-9]+/g, \"-\").replace(/^-|-$/g, \"\"),}, s.heading)\n          , (s.paragraphs || []).map((p, j) => _react2.default.createElement('p', { key: j,}, p))\n          , s.bullets && s.bullets.length ? (\n            _react2.default.createElement('ul', null\n              , s.bullets.map((b, j) => (\n                _react2.default.createElement('li', { key: j,}, b.href ? _react2.default.createElement('a', { href: b.href,}, b.text) : b.text)\n              ))\n            )\n          ) : null\n          , s.note ? _react2.default.createElement('div', { className: \"rr-callout\",}, s.note) : null\n        )\n      ))\n\n      , d.related && d.related.length ? (\n        _react2.default.createElement('section', { 'aria-label': \"Related pages\" ,}\n          , _react2.default.createElement('h2', null, \"Related pages\" )\n          , _react2.default.createElement('ul', null, d.related.map((r, i) => _react2.default.createElement('li', { key: i,}, _react2.default.createElement('a', { href: r.href,}, r.label))))\n        )\n      ) : null\n\n      , d.faqs && d.faqs.length ? (\n        _react2.default.createElement('section', { 'aria-label': \"Frequently asked questions\"  ,}\n          , _react2.default.createElement('h2', null, \"Frequently asked questions\"  )\n          , d.faqs.map((f, i) => (\n            _react2.default.createElement('details', { key: i,}\n              , _react2.default.createElement('summary', null, f.q)\n              , _react2.default.createElement('p', null, f.a)\n            )\n          ))\n        )\n      ) : null\n\n      , _react2.default.createElement('section', { 'aria-label': \"Next step\" ,}\n        , _react2.default.createElement('h2', null, d.closingCtaHeading)\n        , _react2.default.createElement('p', null, d.closingCtaText)\n        , _react2.default.createElement('div', { className: \"rr-ctarow\",}\n          , d.closingCtas.map((c, i) => (\n            _react2.default.createElement('a', { key: i, href: c.href, style: btn(i === 0 ? \"primary\" : \"ghost\"),}, c.label)\n          ))\n        )\n      )\n\n, _react2.default.createElement('style', null, \"\\n.np-rankrift{--np-paper:#eef0ef;--np-ink:#2b3439;--np-accent:#c8ccd9;box-sizing:border-box;max-width:none!important;width:100%;min-height:100vh;background:var(--np-paper)!important;color:var(--np-ink)!important;margin:0!important;padding:0 clamp(20px,5vw,72px) 72px!important;font-family:var(--cdqcat5-font-body, \\\"DM Sans\\\", Arial, sans-serif)!important;font-size:17px!important;line-height:1.75!important;overflow-wrap:anywhere}\\n.np-rankrift *{box-sizing:border-box;min-width:0}.np-rankrift>*{max-width:1080px;margin-left:auto;margin-right:auto}.np-rankrift .np-topbar{display:flex;justify-content:space-between;align-items:center;gap:24px;padding:24px 0;margin-bottom:52px;border-bottom:1px solid var(--np-ink)}.np-rankrift .np-brand{font-family:var(--cdqcat5-font-heading, \\\"Space Grotesk\\\", Arial, sans-serif);font-size:22px;letter-spacing:-.5px;text-decoration:none}.np-rankrift .np-topbar nav{display:flex;flex-wrap:wrap;gap:10px 22px}.np-rankrift .np-topbar nav a{display:inline-flex;align-items:center;min-height:44px;font-size:13px;text-decoration:none;border-bottom:1px solid transparent}.np-rankrift .np-topbar nav a:hover{border-bottom-color:currentColor}\\n.np-rankrift h1{font-family:var(--cdqcat5-font-heading, \\\"Space Grotesk\\\", Arial, sans-serif)!important;font-weight:400!important;font-size:clamp(36px,5.2vw,66px)!important;line-height:1.08!important;letter-spacing:-1.8px!important;max-width:960px;margin-top:0!important;margin-bottom:28px!important;text-wrap:balance;color:var(--np-ink)!important}.np-rankrift h2{font-family:var(--cdqcat5-font-heading, \\\"Space Grotesk\\\", Arial, sans-serif)!important;font-weight:400!important;font-size:clamp(25px,3vw,35px)!important;line-height:1.25!important;letter-spacing:-.5px!important;color:var(--np-ink)!important;margin-top:38px!important;margin-bottom:18px!important;text-wrap:balance}.np-rankrift h3{font-size:19px!important;line-height:1.4!important;color:var(--np-ink)!important;margin-top:24px!important;margin-bottom:10px!important}.np-rankrift p,.np-rankrift li{line-height:1.8!important;max-width:78ch}.np-rankrift p{margin-bottom:18px!important}.np-rankrift p,.np-rankrift li,.np-rankrift label{color:var(--np-ink)!important}.np-rankrift main,.np-rankrift article,.np-rankrift section{max-width:100%}.np-rankrift section{padding-top:8px!important;padding-bottom:24px!important;background:transparent!important}.np-rankrift section+section{border-top:1px solid color-mix(in srgb,var(--np-ink) 22%,transparent);margin-top:32px!important}.np-rankrift ul,.np-rankrift ol{padding-left:24px;max-width:78ch}.np-rankrift li{margin-bottom:10px!important}\\n.np-rankrift a{color:var(--np-ink);text-underline-offset:4px;text-decoration-thickness:1px}.np-rankrift a:hover{text-decoration:underline}.np-rankrift a:focus-visible,.np-rankrift button:focus-visible,.np-rankrift input:focus-visible,.np-rankrift select:focus-visible,.np-rankrift textarea:focus-visible,.np-rankrift summary:focus-visible{outline:3px solid var(--np-ink);outline-offset:5px}.np-rankrift button,.np-rankrift input,.np-rankrift select,.np-rankrift textarea{font:inherit;max-width:100%;min-height:44px;border-radius:3px!important}.np-rankrift button{cursor:pointer}.np-rankrift [class*=\\\"cta\\\"] a,.np-rankrift [class*=\\\"button\\\"]{min-height:44px;display:inline-flex;align-items:center;justify-content:center;padding:12px 20px!important;background:var(--np-ink)!important;color:var(--np-paper)!important;border:1px solid var(--np-ink)!important;border-radius:3px!important;text-decoration:none!important;font-size:14px!important;font-weight:600!important;white-space:normal;line-height:1.45!important}.np-rankrift details{border:1px solid color-mix(in srgb,var(--np-ink) 28%,transparent)!important;border-radius:3px!important;padding:0 18px!important;background:transparent!important;margin-bottom:12px!important}.np-rankrift summary{min-height:48px;padding:14px 0;cursor:pointer;font-weight:600;line-height:1.5}.np-rankrift details>p{padding-bottom:12px}.np-rankrift blockquote{border-left:3px solid var(--np-ink)!important;background:var(--np-accent)!important;color:var(--np-ink)!important;padding:18px 22px!important;margin:24px 0!important;max-width:78ch}.np-rankrift .np-table{width:100%;max-width:100%;overflow-x:auto;overscroll-behavior-x:contain;margin:24px 0;outline-offset:4px}.np-rankrift table{width:100%;border-collapse:collapse;font-size:15px;line-height:1.65;display:table!important;overflow:visible!important}.np-rankrift th{background:var(--np-accent)!important;color:var(--np-ink)!important;text-align:left;padding:15px!important;font-weight:600}.np-rankrift td{color:var(--np-ink)!important;background:transparent!important;padding:15px!important;border-bottom:1px solid color-mix(in srgb,var(--np-ink) 22%,transparent)!important;vertical-align:top}.np-rankrift pre{max-width:100%;overflow-x:auto;padding:20px;background:var(--np-ink)!important;color:var(--np-paper)!important;line-height:1.6;font-size:14px}.np-rankrift pre code{color:inherit!important;background:transparent!important}.np-rankrift img,.np-rankrift svg{max-width:100%;height:auto}\\n@media(max-width:680px){.np-rankrift{font-size:16px!important;padding-left:20px!important;padding-right:20px!important}.np-rankrift .np-topbar{display:block;margin-bottom:36px;padding-top:20px}.np-rankrift .np-topbar nav{margin-top:12px;gap:4px 18px}.np-rankrift h1{letter-spacing:-1px!important}.np-rankrift [style*=\\\"grid-template-columns\\\"]{grid-template-columns:minmax(0,1fr)!important}.np-rankrift [class*=\\\"cta\\\"]{flex-wrap:wrap!important;gap:12px!important}.np-rankrift .np-table table{min-width:580px}.np-rankrift th,.np-rankrift td{padding:12px!important}.np-rankrift blockquote{padding:16px!important}}\\n@media(prefers-reduced-motion:reduce){.np-rankrift *{scroll-behavior:auto!important;transition:none!important}}\\n\")\n, _react2.default.createElement('style', null, \".np-rankrift{--np-paper:var(--cdqcat5r-bg);--np-ink:var(--cdqcat5r-fg);--np-accent:var(--cdqcat5r-panel-muted);--tr-ink:var(--cdqcat5r-fg);--tr-ink-soft:var(--cdqcat5r-fg);--tr-muted:var(--cdqcat5r-muted);--tr-line:var(--cdqcat5r-border);--tr-bg:var(--cdqcat5r-bg);--tr-bg-soft:var(--cdqcat5r-panel-muted);--tr-accent:var(--cdqcat5r-link);--tr-accent-deep:var(--cdqcat5r-link);padding-top:64px!important;padding-bottom:88px!important;font-family:var(--cdqcat5-font-body)!important;background:var(--cdqcat5r-bg)!important;color:var(--cdqcat5r-fg)!important}\\n.np-rankrift h1,.np-rankrift h2,.np-rankrift h3{font-family:var(--cdqcat5-font-heading)!important;font-weight:400!important}.np-rankrift h1{max-width:22ch!important;font-size:clamp(36px,4.8vw,64px)!important;letter-spacing:-.04em!important;line-height:1.12!important}.np-rankrift h2{letter-spacing:-.025em!important}.np-rankrift h3{letter-spacing:-.015em!important}\\n.np-rankrift a{color:var(--cdqcat5r-link)!important}.np-rankrift p,.np-rankrift li{color:var(--cdqcat5r-fg)!important}.np-rankrift [class*=\\\"intro\\\"],.np-rankrift [class*=\\\"lede\\\"]{font-size:20px!important;line-height:1.7!important;max-width:64ch!important;color:var(--cdqcat5r-muted)!important}\\n.np-rankrift [class*=\\\"callout\\\"],.np-rankrift [class*=\\\"note\\\"],.np-rankrift [class*=\\\"summary\\\"],.np-rankrift [class*=\\\"card\\\"],.np-rankrift .tr-cta{background:var(--cdqcat5r-panel-muted)!important;border-color:var(--cdqcat5r-border)!important;color:var(--cdqcat5r-fg)!important;border-radius:8px!important}\\n.np-rankrift [class*=\\\"cta\\\"] a,.np-rankrift [class*=\\\"button\\\"],.np-rankrift .tr-btn{background:var(--cdqcat5r-fg)!important;color:var(--cdqcat5r-bg)!important;border-color:var(--cdqcat5r-fg)!important;min-height:48px!important}.np-rankrift [class*=\\\"eyebrow\\\"]{color:var(--cdqcat5r-link)!important;letter-spacing:.12em!important}.np-rankrift details{border-color:var(--cdqcat5r-border)!important}.np-rankrift th{background:var(--cdqcat5r-panel-muted)!important;color:var(--cdqcat5r-fg)!important}\\n.np-rankrift .tr-hero{padding-top:0!important}.np-rankrift .tr-wrap{max-width:1080px!important;padding-left:0!important;padding-right:0!important}.np-rankrift :focus-visible{outline-color:var(--cdqcat5r-link)!important}@media(max-width:680px){.np-rankrift{padding-top:38px!important;padding-bottom:56px!important}.np-rankrift [class*=\\\"intro\\\"],.np-rankrift [class*=\\\"lede\\\"]{font-size:18px!important}}\")\n)\n  );\n} exports.default = RiskPage;"}