$30M
Launch funding · philanthropic donors including Gates Foundation
2018
A.M. Turing Award — Yoshua Bengio (Nobel Prize of computing)
7+
Board / Advisory Council members — incl. 2 former PMs, Yuval Noah Harari
#1
Most cited living scientist across all fields — Yoshua Bengio
⚠ AUDIT METHODOLOGY NOTE — FIRST NONPROFIT ORG IN THE IDEALAB DATABASE
LawZero is the first nonprofit organization audited using the IdeaLab AI Visibility OS v1.1 framework. All previous audits have been either individual personalities or ecommerce DTC brands. The 7-dimension framework is adapted here for an organizational entity rather than a person: D1 Brand Clarity measures mission coherence and name distinctiveness; D2 Content Depth scores research publications, website content, and press materials; D3 Entity Recognition assesses Wikipedia/Wikidata organizational entity quality; D4 Structured Knowledge evaluates on-site schema and AI crawl signals; D5 Multi-Platform Presence measures social media and media distribution; D6 Social Proof scores donor credibility, board authority, and media citations; D7 AI Discoverability measures how reliably LawZero surfaces in AI responses about AI safety research.
Key scoring difference from personality audits: For a nonprofit research organization, content depth (D2) is weighted toward academic publications and structured research output rather than newsletters or social content. Social proof (D6) is weighted toward institutional donors, board credentials, and peer-cited research rather than follower counts. The entity's AI discoverability (D7) score reflects relevance to research queries and policy discussions rather than consumer product queries.
OVERALL AI VISIBILITY SCORE
AI VISIBILITY
81
/100
B
STRONG ZONE
SCORE SCALE
F0–49Critical Risk
D50–59High Risk
C60–69Moderate
B70–84Strong ← LawZero
A85–100Dominant
DIMENSION SCORES
D4
Structured Knowledge
66
D5
Multi-Platform Presence
62
D6
Social Proof & Citations
94
| DIMENSION | WT | SCORE | GRADE |
|---|
| D1 Brand Clarity | ×1.5 | 86 | A |
| D2 Content Depth | ×1.5 | 72 | B |
| D3 Entity Recognition | ×2.0 | 82 | B |
| D4 Structured Knowledge | ×2.0 | 66 | C |
| D5 Multi-Platform | ×1.5 | 62 | C |
| D6 Social Proof | ×2.0 | 94 | A |
| D7 AI Discoverability | ×2.0 | 80 | B |
| COMPOSITE SCORE | 81 | B |
LawZero is the most consequential nonprofit AI safety research organization launched in 2025 — founded by Yoshua Bengio, the 2018 A.M. Turing Award winner and the most-cited living scientist across all fields. Launched June 3, 2025 with $30 million in philanthropic funding, a board that includes former heads of government (Jacinda Ardern, Stefan Löfven), Yuval Noah Harari, and leaders from Carnegie Endowment and the Nike Foundation, and backed by the Gates Foundation, Schmidt Sciences, Open Philanthropy, and Jaan Tallinn. Its score of 81/100 reflects extraordinary social proof and brand clarity held back by early-stage content infrastructure and nascent social media development — the classic profile of a world-class institution with a world-class team that hasn't yet built world-class owned digital visibility.
GOVERNANCE STRUCTURE — WHY D6 SCORES 94/100
SCIENTIFIC DIRECTOR
Prof. Yoshua Bengio
2018 A.M. Turing Award winner · World's most-cited living scientist across all fields · UdeM Full Professor · Mila Founder
GLOBAL ADVISORY COUNCIL
Dame Jacinda Ardern
Former Prime Minister of New Zealand (2017–2023) · Co-founder, Christchurch Call · Joined March 2026
GLOBAL ADVISORY COUNCIL
Stefan Löfven
Former Prime Minister of Sweden (2014–2021) · AI policy and international governance expertise
BOARD OF DIRECTORS
Yuval Noah Harari
Historian · Bestselling author (Sapiens, Homo Deus, Nexus) · Co-founder, Sapienship social impact company
BOARD CHAIR
Maria Eitel
Founder, Nike Foundation · Founder & Chair Emeritus, Girl Effect · Former VP Corporate Responsibility, Nike
BOARD OF DIRECTORS
Mariano-Florentino Cuéllar
President, Carnegie Endowment for International Peace · Former CA Supreme Court Justice · Stanford Law
PROFILE HIGHLIGHTS
✦ The Bengio entity multiplier — the most powerful founder effect in the database
Yoshua Bengio is the world's most cited living scientist across all fields. His personal Wikipedia entity, academic citation record, and Turing Award status provide LawZero with an entity anchor that no startup of any kind could purchase. When AI systems query "AI safety research organizations" or "safe-by-design AI," Bengio's personal entity pulls LawZero into the response. This is the most powerful single founder effect in the IdeaLab database.
⚡ 13-month-old organization — the gap between institutional authority and owned digital infrastructure
LawZero was founded June 3, 2025. At 13 months old at audit date, it is the youngest entity in the IdeaLab database. The D2 (72) and D5 (62) scores reflect its age — not its ambition or quality. A research organization with this governance and funding typically takes 2–3 years to build the content infrastructure that supports D2 and D5 scores above 80. The trajectory of this audit is exceptional. The current ceiling is time, not strategy.
✔ "Scientist AI" — a proprietary framework with AI training data density
The "Scientist AI" concept — AI that predicts without pursuing goals — is LawZero's core intellectual property and the phrase most likely to be retrieved by AI systems when answering questions about AI safety research. The February 2026 paper "Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?" is published, peer-indexed, and appearing in AI safety discussions. This is LawZero's strongest AI retrieval signal beyond Bengio's personal entity.
7-DIMENSION BREAKDOWN — CLICK TO EXPAND
D1
Brand Clarity
wt ×1.5
STRONGMISSION-DRIVEN CLARITY
86/100
▶
Brand Clarity for an organization measures how unambiguously AI systems can identify its mission, differentiation, and institutional identity. LawZero scores 86/100 — strong, built on an extremely clean mission statement, a distinctive "safe-by-design" positioning, and anchoring to the world's most cited living scientist.
✔"Safe-by-design AI" — a distinctly positioned and attributable mission phrase: The phrase "safe-by-design AI" is used consistently across all LawZero's communications: PRNewswire launch, website, research pages, board appointment press releases, Bloomberg coverage, TED talk, and Yoshua Bengio's personal blog. AI systems encountering this phrase in any context retrieve LawZero with high confidence. The consistency across 13 months of existence is remarkable for a startup organization.
✔"Scientist AI" — proprietary concept creating unique attributable terminology: The Scientist AI — "a mind that has internalized the laws of nature and uses them to make predictions, but without predilection about how things unfold" — is LawZero's most distinctive intellectual contribution. As a named concept appearing in a February 2026 research paper with multiple authors (all associated with LawZero), it creates a citable, attributable, AI-retrievable concept node. AI systems discussing AI safety research increasingly reference the Scientist AI framing.
✔Bengio anchoring creates unambiguous entity context: "LawZero, the nonprofit founded by Yoshua Bengio" is the most common sentence pattern in press coverage of the organization. Bengio's entity — Turing Award, most-cited living scientist, Mila founder — immediately contextualises LawZero for any AI system. The entity disambiguation is near-zero: no other organization is called LawZero, and no other AI safety nonprofit has this founder profile.
△Name disambiguation risk — "Law Zero" could be misread as legal/law sector: The name "LawZero" (without context) could be disambiguated as a legal technology company, a law firm, or a legal AI product. In isolation, "LawZero.org" does not immediately signal AI safety research to a first-time reader or a cold-context AI query. The "Scientist AI" and "Yoshua Bengio" anchors solve this in context, but Organization JSON-LD with explicit knowsAbout would eliminate the ambiguity structurally.
△Bilingual site (EN/FR) — distributed entity signal: lawzero.org operates in both English and French (reflecting Montréal base). While this is appropriate and positive for audience reach, the bilingual URL structure (/en/ and /fr/) distributes link authority and AI crawl signals across two path prefixes. Canonical URL declaration and hreflang implementation are important to ensure AI crawlers attribute both to the same entity.
✦ UNIQUE ADVANTAGE
LawZero is the only entity in any IdeaLab audit where the founder's personal entity is so strong that it effectively provides the organization with a Wikipedia-level entity anchor by proxy. Bengio's Wikipedia page mentions LawZero explicitly — creating a bidirectional entity link between the most-cited scientist in the world and LawZero as an organization.
→ ACTION
Add Organization JSON-LD to lawzero.org homepage with knowsAbout: ["AI safety", "safe-by-design AI", "Scientist AI", "AI alignment", "AI governance"] and founder → Person (Yoshua Bengio, linking to his Wikidata/Wikipedia entity). This structurally resolves the name disambiguation risk.
D2
Content Depth
wt ×1.5
GROWING13 MONTHS OLD
72/100
▶
Content Depth for a research organization measures the volume, specificity, and crawlability of owned research publications, website content, press releases, and thought-leadership material. LawZero scores 72/100 — strong for a 13-month-old nonprofit, reflecting a lean but high-quality content corpus that is growing rapidly.
✔Research publications page with indexed academic content: lawzero.org/en/research hosts multiple indexed research publications. The February 2026 paper "Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?" is the flagship publication — authored by Bengio plus 12 co-researchers, published on lawzero.org and likely submitted to arXiv. This single paper generates AI training data signals across multiple indexed repositories simultaneously.
✔News section with structured press releases (PRNewswire-distributed): LawZero's news/releases section documents the June 2025 launch, January 2026 board announcement, and March 2026 Jacinda Ardern appointment — all distributed via PRNewswire (a high-DA wire service that AI training pipelines index extensively). Each PRNewswire release creates an indexed copy at multiple syndication destinations simultaneously.
✔Yoshua Bengio's personal blog (yoshuabengio.org) — extended content layer: The "Introducing LawZero" post at yoshuabengio.org/2025/06/03/introducing-lawzero/ creates a high-DA personal blog post from the most cited scientist in the world, explicitly describing LawZero's mission, rationale, and scientific direction. This external content extends LawZero's indexed text corpus significantly beyond the owned domain.
△Limited editorial content beyond research and press releases: For a nonprofit with this mandate, there is significant opportunity to publish accessible explainers of the Scientist AI concept, blog posts on AI safety developments, op-ed commentary on AI governance, and accessible summaries of research findings. These would dramatically expand the AI-indexable text corpus in query-relevant formats without requiring additional research output.
✗Research papers not yet on arXiv (or insufficient arXiv citation volume): The strongest AI training data signal for a research organization is arXiv pre-print listings, which are indexed by AI training pipelines globally. If LawZero's research is not on arXiv or similar open-access repositories (beyond the lawzero.org/research page), it is missing the most important academic content distribution channel for AI retrieval.
◆ GAPS
Limited editorial content (explainers, accessible science writing, op-eds)Research papers may not all be on arXiv — most AI training pipelines weight arXiv heavilyNo newsletter or regular subscription-based content publication→ ACTIONS
Ensure all research papers are submitted to arXiv with lawzero.org DOILaunch a monthly "AI Safety Research Update" newsletter (Substack or direct)Publish 6–8 accessible blog posts explaining Scientist AI, contextualization, consequence invariance for non-academic audiences
D3
Entity Recognition
wt ×2.0
STRONGBENGIO ENTITY ANCHOR
82/100
▶
Entity Recognition for an organization measures how well AI knowledge graphs have mapped the entity — Wikipedia organizational page, Wikidata entry, Crunchbase listing, and the bidirectional links to named founders and board members who already have established entity graphs. LawZero's 82/100 is the highest D3 score of any entity in the database that lacks its own Wikipedia article, entirely due to the Bengio entity proxy effect.
✔Yoshua Bengio Wikipedia entity — the most powerful proxy anchor in the database: Bengio's Wikipedia page is one of the most detailed AI researcher profiles on Wikipedia globally. It mentions LawZero explicitly as his current primary role ("Co-President and Scientific Director, LawZero"). For any AI system querying Bengio's entity, LawZero appears as his primary organizational affiliation — creating a bidirectional entity link that most organizations could not purchase at any price.
✔Incubated at Mila — institutional affiliation with a Wikipedia-listed research institute: Mila – Quebec AI Institute has its own Wikipedia entity, is listed in AI research databases globally, and has Bengio as its founder. LawZero's incubation at Mila creates an institutional entity link chain: LawZero → Mila (Wikipedia entity) → Bengio (Wikipedia entity). This chain gives LawZero entity recognition well above its 13-month age would typically support.
✔Board members with independent Wikipedia entities (Harari, Ardern, Löfven, Cuéllar): Every major figure on LawZero's board has their own Wikipedia entity — Yuval Noah Harari (globally recognized author), Jacinda Ardern (former PM), Stefan Löfven (former PM), Mariano-Florentino Cuéllar (former CA Supreme Court Justice). Each of these entities, when updated to include "member of LawZero's board," creates an additional entity link pointing to LawZero.
△No dedicated LawZero Wikipedia article yet: Despite the extraordinary governance roster and $30M in funding from institutional donors, LawZero does not yet have its own Wikipedia organizational article. At 13 months old, this is not unexpected — Wikipedia's notability guidelines for organizations typically require sustained coverage over time. By June 2026 (12+ months of press coverage from PRNewswire, Bloomberg, The Next Web, UdeM Nouvelles), the notability case is almost certainly met.
△Wikidata organizational entity status unclear: It is not confirmed whether LawZero has a Wikidata organizational item. If not, creating one (with P31: nonprofit organization, P571: founded June 2025, P127: Yoshua Bengio, P856: lawzero.org, P17: Canada) would immediately create a machine-readable entity node in the world's primary open knowledge graph.
◆ GAPS
No dedicated LawZero Wikipedia article (notability criteria met — eligible to submit)Wikidata organizational entity status uncertainBoard members' Wikipedia entities may not yet list LawZero as an affiliation→ ACTIONS
Submit LawZero Wikipedia organization article — PRNewswire + Bloomberg + TED coverage provide clear notability evidenceCreate/verify Wikidata organizational entity for LawZeroRequest board member Wikipedia editors to add LawZero to their affiliation sections
D4
Structured Knowledge
wt ×2.0
WATCHSCHEMA GAPS
66/100
▶
Structured Knowledge scores on-page schema markup completeness, AI crawl signals, and the machine-readability of key organizational facts. LawZero scores 66/100 — the primary gap dimension, reflecting the early-stage nature of the website's technical SEO infrastructure. A research organization with this pedigree has specific schema opportunities unavailable to consumer brands.
✔Research publication pages with structured academic content: The lawzero.org/en/research page hosts publications with titles, authors, and dates — the basic metadata that AI crawlers index for academic entity attribution. This is above-average for a nonprofit website and reflects the organization's academic DNA.
✔PRNewswire wire service distribution creates external structured data: Every PRNewswire release creates hundreds of syndicated copies with consistent metadata (organization name, date, location, contact). This wire service distribution is a form of distributed structured data that AI training pipelines index — it is why D4 is 66 rather than lower despite on-site schema gaps.
✗No Organization JSON-LD on homepage: lawzero.org does not appear to deploy explicit Organization schema declaring the nonprofit type, founding date, founder, funding sources, research areas, and sameAs links. For an organization whose mission directly concerns AI information systems, deploying structured data that AI crawlers can read is both strategically obvious and practically straightforward.
✗No LLMs.txt: lawzero.org does not deploy an LLMs.txt file, meaning AI crawlers (GPTBot, ClaudeBot, PerplexityBot) cannot know which pages the organization considers most authoritative. For an AI safety research organization, having an explicit AI crawl manifest is both practically valuable and symbolically resonant — LawZero builds AI systems and should be exemplary in how it communicates with AI crawlers.
✗Research papers lack Article/ScholarlyArticle JSON-LD: The research publications at lawzero.org/en/research do not appear to carry ScholarlyArticle JSON-LD (author, datePublished, abstract, keywords, isPartOf). This schema type is specifically designed for academic content and is indexed by Google Scholar, Semantic Scholar, and AI training pipelines that weight academic attribution.
◆ GAPS (priority order)
No Organization JSON-LD on homepage — highest-ROI single fixNo LLMs.txt — symbolically important for an AI safety orgNo ScholarlyArticle JSON-LD on research publicationsBilingual site canonical/hreflang implementation needs verification→ ACTIONS
Deploy Organization JSON-LD: nonprofit type, founding date, Bengio as founder, Gates Foundation/Schmidt as funders, knowsAbout AI safety topics, sameAs WikidataCreate /llms.txt — list research pages, team pages, and about page as AI priority contentAdd ScholarlyArticle JSON-LD to each research paper page
D5
Multi-Platform Presence
wt ×1.5
NASCENTGROWTH OPPORTUNITY
62/100
▶
Multi-Platform Presence for a nonprofit research organization measures its distribution across channels where AI models train and retrieve — including LinkedIn, X, YouTube, academic databases, and media appearances. LawZero scores 62/100 — reflecting early-stage social media development appropriate to its age, with significant upside as the organization matures.
✔LinkedIn organizational presence (confirmed): LawZero has an active LinkedIn company page, which is the primary B2B platform for research organization communication. For talent recruitment, donor relations, and policy community engagement — the three core stakeholder audiences of a nonprofit AI safety lab — LinkedIn is the correct primary social platform.
✔TED2025 appearance — "Humanity Reimagined" talk by Bengio: Yoshua Bengio's TED2025 talk, explicitly linked from the LawZero homepage, extends the organization's platform presence to one of the world's most authoritative thought leadership stages. TED talk content is indexed, transcribed, and cited by AI systems across multiple platforms — creating a high-quality media appearance that functions as distributed content for LawZero's mission.
✔Bloomberg feature (June 15, 2026) — "The catastrophic risks of AI — and a safer path": A Bloomberg feature published 19 days before this audit is the most recent major media event. Bloomberg features are indexed by AI training pipelines with high weight due to Bloomberg's domain authority and financial/technology audience. This is a recency signal that AI systems weight for freshness in AI safety queries.
△Social media following nascent across all platforms: At 13 months old with a research-first orientation, LawZero's social following is modest relative to its institutional stature. This is expected and appropriate — a nonprofit AI safety lab's primary audience is researchers, policymakers, and philanthropists, not general consumers. However, increasing publication of accessible content (explainers, research summaries, policy commentary) would grow platform presence in the query categories where AI systems retrieve LawZero most frequently.
✗No YouTube channel confirmed: For an AI safety research organization with a TED speaker, a research team giving conference talks, and a mandate to communicate risks to the public, YouTube is a critical missing platform. Video content on YouTube creates audio-transcribed, AI-indexed content at scale. A YouTube channel with accessible explanations of the Scientist AI concept, monthly research updates, and policy commentary would significantly expand D5 and D2 simultaneously.
◆ GAPS
No YouTube channel for accessible research communicationSocial following nascent — not yet a research distribution assetNo Bluesky/academic social (important for AI safety researcher community)→ ACTIONS
Launch YouTube channel — upload TED talk, create monthly "AI Safety Research" explainer seriesEstablish Bluesky presence for AI safety community (significant portion of AI safety researchers are on Bluesky)Cross-post research summaries to X/Twitter where Bengio's personal account amplifies them
D6
Social Proof & Citations
wt ×2.0
ELITE · 94/100 · TIED FOR HIGHEST IN DATABASE
94/100
▶
Social Proof & Citations for a nonprofit organization measures donor credibility, board authority, media coverage quality, and peer recognition. LawZero scores 94/100 — tied with Lily Ray for the highest D6 in the entire IdeaLab database, including both personality and ecommerce audits. This score is built on arguably the most institutional citation portfolio assembled by any entity audited to date.
✔Gates Foundation as donor — the highest-credibility philanthropic citation on Earth: The Bill & Melinda Gates Foundation is the largest private charitable foundation in the world. Its institutional citation is recognised by AI systems as a tier-1 credibility signal for any nonprofit. Being listed as a Gates Foundation-supported organization in PRNewswire press releases creates a permanent, structured citation that AI training pipelines weight as maximum institutional credibility.
✔Schmidt Sciences, Open Philanthropy, Future of Life Institute, Jaan Tallinn — the AI safety philanthropic tier-1 list: LawZero's donor list ($30M at launch) reads like a directory of the world's most credible AI safety funders. Each of these donors has their own entity recognition, press coverage, and AI training data presence — and their association with LawZero creates a citation cluster that AI systems recognise as the gold standard of AI safety research legitimacy.
✔Turing Award founder + two former PMs + Sapiens author on advisory structure: No other organization in the IdeaLab database — across all categories — has a governance roster that matches this. Yoshua Bengio (most-cited living scientist, Turing Award), Jacinda Ardern (former PM of NZ, Christchurch Call co-founder), Stefan Löfven (former PM of Sweden), Yuval Noah Harari (Sapiens, 20M+ copies sold), Mariano-Florentino Cuéllar (Carnegie Endowment President, former CA Supreme Court Justice), Maria Eitel (Nike Foundation founder). This is a social proof portfolio that defies category comparison.
✔Bloomberg June 2026, TED2025, PRNewswire ×3, The Next Web, UdeM Nouvelles, Wall Street Journal (via Bengio interview): The press portfolio is appropriately institutional — the publications that matter for LawZero's target audiences (AI researchers, policymakers, philanthropists, tech executives) are all represented. Bloomberg and TED are particularly high-weight citations for AI systems answering questions about AI safety and governance.
✦ DATABASE RECORD
D6 of 94/100 tied with Lily Ray for the highest Social Proof score in the IdeaLab database. The combination of Gates Foundation donor citation, Turing Award founder, two former PM advisors, and Sapiens author board member creates a citation portfolio with no equivalent in the database.
→ ACTION
Encode all donor and award citations in Organization JSON-LD: funder array (Gates Foundation, Schmidt Sciences, Open Philanthropy, Jaan Tallinn/SVCF, Future of Life Institute), member (board members with Wikipedia sameAs links), foundingDate: "2025-06-03". This moves citations from press-crawlable to structured-data-readable.
D7
AI Discoverability
wt ×2.0
STRONGQUERY-CATEGORY DOMINANT
80/100
▶
AI Discoverability measures how reliably LawZero surfaces in AI-generated responses across its target query categories. LawZero scores 80/100 — strong for a 13-month-old organization, built primarily on the Bengio entity anchor and the Scientist AI concept's growing presence in AI safety discourse.
✔Dominant in "Yoshua Bengio AI safety" and "Scientist AI" queries: Any AI system querying Bengio's current work will return LawZero as his primary institutional affiliation. The Scientist AI paper and concept are increasingly appearing in AI safety research discussions across academic and policy platforms. This query-category dominance is built on the strongest possible pre-training data foundation — a Turing Award winner's primary affiliation.
✔AI safety research organization category — growing retrieval frequency: The AI safety research category is one of the fastest-growing query categories in AI systems in 2025–2026. As public awareness of AI risks grows (reflected in Bloomberg's June 2026 coverage and The Next Web's May 2026 piece on Bengio's extinction warning), query volume for "nonprofit AI safety organizations" is increasing. LawZero's early positioning in this category — with the most credible governance structure — gives it a compounding AI retrieval advantage.
✔TED talk and Bloomberg feature create RAG-retrievable content: Unlike social media content (which may not be indexed by RAG systems), TED talk transcripts and Bloomberg articles are directly retrieved by AI search engines (Perplexity, ChatGPT Search) when answering AI safety queries. These two pieces of media content function as active RAG citations for LawZero — not just pre-training data.
△Narrowly discoverable — queries must include Bengio or AI safety specifics: LawZero's current discoverability is strongest when queries include "Yoshua Bengio," "Scientist AI," or "nonprofit AI safety research." Broader queries like "AI safety organizations," "who is working on safe AI," or "alternatives to OpenAI for AI safety" may not yet reliably surface LawZero. As owned content depth grows, broader query coverage will expand.
✗No LLMs.txt — AI safety research organization irony: LawZero builds AI systems and studies AI behavior. Not having an LLMs.txt file on lawzero.org is an organizational alignment gap — the organization that is studying how AI crawlers should behave has not deployed the tool that communicates intent to AI crawlers on its own domain.
✦ GROWTH TRAJECTORY
D7 will improve organically as content depth grows and as the AI safety query category expands. LawZero is positioned to be the dominant research organization in AI-generated AI safety responses within 18–24 months if content infrastructure is built. The Bengio entity anchor makes this trajectory near-certain for query categories that include his name.
→ ACTIONS
Deploy /llms.txt immediately — the symbolism of an AI safety org having this is notablePublish accessible explainer content targeting broader AI safety query patternsEnsure all research papers are on arXiv — AI research query engines heavily weight arXiv
CONFIRMED STRENGTHS vs. CONFIRMED GAPS
✔ CONFIRMED STRENGTHS
✔Yoshua Bengio Wikipedia entity — world's most-cited living scientist, LawZero listed as primary affiliation
✔Gates Foundation donor citation — tier-1 institutional credibility signal globally
✔Schmidt Sciences, Open Philanthropy, Jaan Tallinn/SVCF, Future of Life Institute donors
✔$30M launch funding — verifiable, multi-source cited financial credibility
✔Jacinda Ardern + Stefan Löfven on advisory council — former PM governance authority
✔Yuval Noah Harari + Maria Eitel + Cuéllar on board — cross-disciplinary institutional authority
✔PRNewswire distribution ×3 — high-DA wire service syndication creating structured citations
✔TED2025 "Humanity Reimagined" — Bengio talk explicitly for LawZero mission
✔Bloomberg June 2026 feature — recency signal on highest-authority tech/finance media
✔"Scientist AI" concept — proprietary named framework with February 2026 research paper
✔Incubated at Mila (Wikipedia entity) — institutional origin creating entity link chain
✔Bilingual site EN/FR — appropriate for Montréal-based international organization
✗ CONFIRMED GAPS
✗No dedicated LawZero Wikipedia organization article (notability criteria met)
✗No Organization JSON-LD on homepage — entity disambiguation and donor citation not machine-readable
✗No LLMs.txt — symbolically and practically important for an AI safety research org
✗No ScholarlyArticle JSON-LD on research publication pages
✗No YouTube channel for research communication and public engagement
✗Research papers may not all be on arXiv — critical for AI research query coverage
✗No newsletter for regular research updates (Substack or direct)
✗Wikidata organizational entity status unconfirmed
✗Board member Wikipedia entities may not yet list LawZero as affiliation
✗Accessible science blog for non-academic audiences (Scientist AI explainers)
✗Bluesky/academic social presence for AI safety research community
PRIORITISED ACTION PLAN — PATH FROM 81 TO 90+
P1
D4 Schema
Deploy Organization JSON-LD on lawzero.org homepage
+7–9 pts on D4 · +4 on D1
2–3 hours · highest single-action ROI
Add a comprehensive Organization JSON-LD block to the lawzero.org homepage (and its /en equivalent). Required properties: @type: "ResearchOrganization", name: "LawZero", alternateName: "Law Zero", url: "https://lawzero.org", foundingDate: "2025-06-03", foundingLocation (Montréal, QC, Canada), founder (Yoshua Bengio, with sameAs linking to his Wikipedia and Wikidata entities), funder (array: Gates Foundation, Schmidt Sciences, Open Philanthropy, Future of Life Institute, Jaan Tallinn/SVCF), member (board + advisory council with Wikipedia sameAs links for each), knowsAbout (AI safety, safe-by-design AI, Scientist AI, AI alignment, AI governance, frontier AI risk, deep learning), sameAs (Wikidata entity URL, LinkedIn company page URL). This single action moves all of LawZero's extraordinary institutional credentials from "discoverable by crawling press releases" to "machine-readable from the organization's own domain."
P2
D4 + D7
Create /llms.txt — the AI safety org's AI crawl declaration
+3–4 pts on D7
1 afternoon · highest symbolic + practical value
Deploy lawzero.org/llms.txt listing the pages AI crawlers should prioritise: the research publications page (lawzero.org/en/research), the Scientist AI description page, the About/team page, the news archive, and the organization's primary English homepage. Include brief descriptions of each page's purpose and content type. This action is particularly symbolically resonant for LawZero: the organization that builds AI systems and studies AI behavior should be exemplary in how it communicates with AI crawlers. It is also practically important — without LLMs.txt, GPTBot and PerplexityBot index by probability rather than editorial intent, potentially prioritising the French-language pages or older news items over the research content that defines LawZero's authority.
P3
D3 Entity
Submit LawZero Wikipedia organization article
+12–16 pts on D3 · cascades to D7
4–6 weeks · notability criteria clearly met
LawZero's Wikipedia notability case is stronger than any organization audited by IdeaLab to date. References for the article: (1) PRNewswire June 3, 2025 launch release — Tier 1 wire service, independently distributed. (2) PRNewswire January 15, 2026 board announcement. (3) PRNewswire March 5, 2026 Jacinda Ardern appointment. (4) The Next Web May 2026 — "Yoshua Bengio warns hyperintelligent AI with preservation goals could threaten human extinction." (5) Bloomberg June 15, 2026 — "The catastrophic risks of AI — and a safer path." (6) UdeM Nouvelles June 3, 2025 — "Safe-by-design AI: Yoshua Bengio launches LawZero." The article should cover: founding context (response to frontier AI risks), scientific direction (Scientist AI), governance structure, donor portfolio, and current research. Draft in neutral encyclopedic tone, third-person, no promotional language. Wikipedia's threshold for nonprofit organizations is met with 3+ independent, reliable secondary sources — LawZero has 6+ from internationally recognized press outlets.
P4
D2 Content
Add ScholarlyArticle JSON-LD to all research pages + submit all papers to arXiv
+5–7 pts on D2 · +3 pts on D4
2–4 weeks · academic infrastructure sprint
Two actions in parallel: (1) Add ScholarlyArticle JSON-LD to each publication on lawzero.org/en/research — properties: headline (paper title), author (array of Person entities with name + affiliation), datePublished, abstract, keywords (AI safety, Scientist AI, etc.), publisher (Organization: LawZero), isAccessibleForFree: true. This makes LawZero's research Google Scholar-eligible and creates structured academic metadata that AI research query engines use for attribution. (2) Ensure all research papers are submitted to arXiv.org with LawZero affiliation marked for all authors. arXiv is the primary open-access research repository indexed by AI training pipelines globally. Every paper on arXiv creates a permanent, structured academic citation with author names, affiliations, abstracts, and keywords — all machine-readable and AI-training-weighted.
P5
D2 + D5
Launch YouTube channel + accessible science blog
+5–8 pts on D5 · +4 pts on D2
1–3 months · content ops investment
Two content infrastructure initiatives: (1) YouTube channel — upload Yoshua Bengio's TED2025 talk with permission, create a recurring "Scientist AI Explained" series (5–10 minute accessible videos explaining core concepts: what is the Scientist AI, what is consequence invariance, how does contextualization work, why agency is the key risk in frontier AI). Each video's auto-generated transcript is indexed by Google and AI training pipelines. (2) Accessible science blog — publish 8–10 accessible explainer posts on lawzero.org/blog targeting natural-language query patterns: "What is the Scientist AI?", "Why is agentic AI dangerous?", "What does 'safe-by-design AI' mean?", "How is LawZero different from other AI safety labs?". These posts create the owned editorial content depth that research publications alone cannot provide — reaching policy, philanthropy, and general audiences that don't read academic papers.
P6
D3 Entity
Update board member Wikipedia entities to include LawZero affiliation
+4–6 pts on D3
Coordinate with board + Wikipedia community
Jacinda Ardern, Yuval Noah Harari, Stefan Löfven, Mariano-Florentino Cuéllar, and Maria Eitel all have Wikipedia entities. If none of these entities currently list LawZero as an organizational affiliation, updating them would create 5 additional Wikipedia entity links pointing to LawZero — multiplying the entity graph connections available to AI knowledge graph queries. This is best coordinated through the Wikipedia community (not directly editing board members' own pages), using the PRNewswire press releases as reference citations. Each board appointment press release is a citable Wikipedia reference — the January 15 and March 5, 2026 PRNewswire releases are exactly the type of independent, reliable secondary sources Wikipedia editors use to add organizational affiliations.
SCORE PROJECTION ROADMAP
WEEK 1–4 · SCHEMA + ENTITY SPRINT
81 → 87 projected
✔ Organization JSON-LD on homepage
✔ LLMs.txt deployed
✔ ScholarlyArticle JSON-LD on research pages
✔ Wikidata organizational entity created
✔ arXiv submission for all papers
✔ Board Wikipedia affiliations updated
MONTH 1–3 · WIKIPEDIA + CONTENT
87 → 92 projected
✔ LawZero Wikipedia article submitted + approved
✔ YouTube channel launched
✔ 8 accessible science blog posts published
✔ Research newsletter launched
✔ Bluesky presence established
✔ Dominant in AI safety research AI queries
MONTH 6–18 · AUTHORITY GROWTH
92 → 96+ projected
✔ 5+ research papers published + arXiv cited
✔ Nature / Science feature coverage
✔ Government policy citations emerging
✔ UN / OECD AI safety citations
✔ YouTube 10K+ subscribers (research audience)
✔ Dominant AI safety org in all AI query contexts
BENCHMARK — LAWZERO vs. IDEALAB DATABASE (CROSS-CATEGORY)
Lily Ray · SEO / AI Search Expert (Personality)84
Bellroy (bellroy.com) · Ecommerce DTC83
LawZero (lawzero.org) · Nonprofit AI Safety ◀ THIS AUDIT81
Peak Design (peakdesign.com) · Ecommerce DTC79
Dan Koe · Creator Economy (Personality)77
Alex Hormozi · Business / Offer Creation (Personality)87
* Cross-category comparison — personalities, ecommerce brands, and nonprofit organizations scored on same IdeaLab AI Visibility OS v1.1 framework. At 81/100, LawZero ranks above Peak Design and Dan Koe despite being 13 months old — driven by the extraordinary D6 (94/100, tied for highest in database) and the Bengio entity proxy effect on D3. Post-schema sprint and Wikipedia article, LawZero projects to 92 — the highest organizational score in the database, surpassing even Hormozi (87). The trajectory of a 13-month-old nonprofit with $30M in funding and the world's most-cited scientist as founder is unlike anything else in this database.
AUDIT VERDICT
LawZero scores 81/100 — Grade B, Strong Zone — at 13 months old, the highest-trajectory score in the IdeaLab database. LawZero is the first nonprofit organization audited in the IdeaLab AI Visibility OS framework, and it presents the most unusual scoring profile in the database: extraordinary social proof (D6: 94/100, tied for highest ever recorded) and strong entity recognition (D3: 82, the highest for any entity without its own Wikipedia article) — combined with content depth and platform presence that reflect an organization that has existed for barely a year. The gap between LawZero's institutional authority and its owned digital infrastructure is the widest of any entity audited, and it will close rapidly and naturally as the organization grows.
The defining finding of this audit is the Bengio entity proxy effect. Yoshua Bengio is the most cited living scientist across all fields — more cited than any physicist, biologist, economist, or computer scientist alive. His Wikipedia entity is comprehensive, his Turing Award is the highest honour in computing, and his name is associated with LawZero as Co-President and Scientific Director in every press release, news article, and academic citation. For AI systems retrieving information about AI safety research, Bengio's entity is the most powerful possible founder anchor — it gives LawZero entity recognition and AI discoverability that most organizations cannot achieve in a decade.
The path from 81 to 92 is a schema sprint, a Wikipedia article, and an arXiv submission process — the kind of technical infrastructure work that a team with LawZero's capabilities can execute in 4–6 weeks. At 92+, LawZero would be the highest-scoring entity in the entire IdeaLab database across all categories — personalities, ecommerce, and nonprofit. The only entity in the database with the social proof, entity authority, and institutional credibility to project that score at 13 months old is the organization founded by the world's most-cited living scientist to build AI systems that are safe by design. The irony of course is that LawZero — the organization building AI that understands the world but does not act on goals of its own — has not yet deployed the structured data signals that tell AI systems about its own goals. That is the fix.
AUDIT DATABASE RECORD
| entity_name | LawZero |
| primary_domain | lawzero.org |
| entity_type | Nonprofit organization · AI safety research · First nonprofit in IdeaLab database |
| location | Montréal, Québec, Canada · Incubated at Mila – Quebec AI Institute |
| founded | June 3, 2025 · Yoshua Bengio · 13 months at audit date |
| audit_date | 2026-07-04 · IdeaLab AI Visibility OS v1.1 |
| audit_type | Nonprofit Org Audit · First in category · Framework adapted for organizational entity |
| composite_score | 81 / 100 |
| grade | B — Strong Zone · Highest trajectory score in database at 13 months old |
| D1_brand_clarity | 86 · A — "Safe-by-design AI" consistent mission phrase. "Scientist AI" proprietary concept. Bengio anchor. Name disambiguation risk managed by context. |
| D2_content_depth | 72 · B — Research publications page, PRNewswire ×3, Bengio's personal blog. Limited editorial content. Age-appropriate score with strong growth trajectory. |
| D3_entity_recognition | 82 · B — Bengio Wikipedia entity lists LawZero as primary affiliation. Mila entity link. Board members with Wikipedia entities. No dedicated LawZero article yet. |
| D4_structured_knowledge | 66 · C — No Organization JSON-LD. No LLMs.txt. No ScholarlyArticle schema. PRNewswire syndication provides distributed structured data. |
| D5_multiplatform | 62 · C — LinkedIn confirmed. TED2025 appearance. Bloomberg June 2026. No YouTube. Nascent social. Age-appropriate profile. |
| D6_social_proof | 94 · A — TIED FOR HIGHEST IN DATABASE. Gates Foundation + Schmidt Sciences + Open Philanthropy + Jaan Tallinn. Jacinda Ardern + Löfven + Harari + Eitel + Cuéllar on governance. $30M, TED, Bloomberg. |
| D7_ai_discoverability | 80 · B — Dominant in Bengio-anchored AI safety queries. Scientist AI concept growing. Bloomberg + TED as active RAG sources. No LLMs.txt. Broadening query coverage needed. |
| founder_entity_effect | Yoshua Bengio — most-cited living scientist globally (all fields). 2018 A.M. Turing Award. Wikipedia entity comprehensive, lists LawZero as primary affiliation. Most powerful single founder effect in IdeaLab database. |
| unique_asset | Gates Foundation donor citation — highest credibility philanthropic signal available. Combined with Turing Award founder: no entity in IdeaLab database has equivalent institutional authority anchor. |
| key_gap | D4 Structured Knowledge (66) — Organization JSON-LD, LLMs.txt, ScholarlyArticle schema all absent. All three are deployable within one afternoon. |
| projected_week4 | 87 / 100 — after Organization JSON-LD + LLMs.txt + ScholarlyArticle schema + arXiv submissions + Wikidata entity |
| projected_90days | 92 / 100 — after Wikipedia article + YouTube + accessible blog + newsletter. Highest organizational score in database. |
| projected_18months | 96+ / 100 — after research publication volume grows, policy citations emerge, Nature/Science coverage develops |
| auditor | IdeaLab.ai · AI Visibility OS v1.1 · idea-lab.ai/audits |
IS YOUR ORGANIZATION SCORED IN THE IDEALAB AI VISIBILITY DATABASE?
Get your AI Visibility Score in 48 hours.
The IdeaLab AI Visibility OS v1.1 audits brands, personalities, and organizations across 7 dimensions — delivering a prioritised action plan that starts with the changes adding the most points in the least time.