DRAFT v0.1 — FOR ATTORNEY REVIEW ONLY. NOT LEGAL ADVICE. NOT YET EFFECTIVE.
Last updated: [EFFECTIVE DATE] · Version: 0.1-DRAFT
*This AI Disclosure & Disclaimer (this "Disclaimer") is published by Cannvas, LLC (d/b/a "Cannvas," "we," "us," or "our"). It supplements, and is incorporated by reference into, (a) the Master SaaS Subscription Agreement governing business/operator use of the Service (the "MSA") and (b) the Consumer Terms of Use governing use of the Consumer Apps (the "Consumer Terms"). In the event of a conflict between this Disclaimer and the MSA or Consumer Terms as to the subject matter addressed here, this Disclaimer controls with respect to AI Outputs; in all other respects the MSA or Consumer Terms (as applicable) control. Capitalized terms used but not defined here have the meanings given in the MSA or Consumer Terms.*
1.1 Purpose. Cannvas provides ancillary cannabis-industry software. Certain features of the Service use artificial intelligence, machine learning, statistical models, rules engines, and other automated or algorithmic techniques (collectively, "AI/Automated Systems") to generate outputs, estimates, recommendations, classifications, scores, projections, narratives, and insights (collectively, "AI Outputs"). This Disclaimer explains, at a high level, (a) that and how AI/Automated Systems are used, (b) the material limitations of AI Outputs, and (c) the responsibilities of persons who view or act on AI Outputs.
1.2 Who this applies to. This Disclaimer applies to every Organization, Authorized User, and Consumer that accesses or uses any part of the Service where AI Outputs are surfaced, including the Dashboard and the Consumer Apps.
1.3 Scope. This Disclaimer addresses AI Outputs only. It does not modify the parties' respective rights and obligations under the MSA or Consumer Terms except as expressly stated in the preamble above. This Disclaimer is not a warranty, service-level commitment, or guarantee of any result.
For purposes of this Disclaimer:
3.1 Feature inventory. AI/Automated Systems are used to generate, assist in generating, rank, narrate, or otherwise inform AI Outputs across the Service, including without limitation:
3.2 Non-exhaustive; subject to change. The list in Section 3.1 is illustrative and non-exhaustive. Features may be added, changed, or removed, and additional parts of the Service may use AI/Automated Systems over time. Where practicable, the Service identifies AI Outputs at or near the point of use (see Section 10).
4.1 AI-assisted and probabilistic. AI OUTPUTS ARE GENERATED, IN WHOLE OR IN PART, BY AI/AUTOMATED SYSTEMS. These systems are PROBABILISTIC AND STATISTICAL IN NATURE, not deterministic oracles of truth. They produce estimates, likelihoods, and pattern-based inferences that can vary and that may be presented in natural-language ("narrated") form.
4.2 Reliance on historical, incomplete, or estimated data. AI Outputs may draw on HISTORICAL, INCOMPLETE, ESTIMATED, DELAYED, THIRD-PARTY, OR MODELED DATA (including data synchronized from or dependent on third-party systems such as compliance/track-and-trace and laboratory sources). Past performance and historical patterns do not guarantee future results. Data latency, gaps, mapping ambiguities, and upstream errors can affect AI Outputs.
4.3 Third-party models and components. The Service may use third-party AI models, libraries, and infrastructure. Cannvas does not control, and does not represent or warrant the behavior of, third-party models, and the involvement of any third-party component does not enlarge Cannvas's obligations under this Disclaimer, the MSA, or the Consumer Terms.
5.1 NO GUARANTEE OF ACCURACY. AI OUTPUTS ARE PROVIDED "AS IS" AND "AS AVAILABLE" AND MAY BE INACCURATE, INCOMPLETE, OUTDATED, MISLEADING, OR WRONG. AI OUTPUTS MAY CONTAIN ERRORS, OMISSIONS, MISCLASSIFICATIONS, MISCALCULATIONS, OR FABRICATED, UNSUPPORTED, OR NONSENSICAL CONTENT (COMMONLY CALLED "HALLUCINATIONS"). CANNVAS DOES NOT WARRANT OR GUARANTEE THAT ANY AI OUTPUT IS ACCURATE, COMPLETE, CURRENT, RELIABLE, FIT FOR A PARTICULAR PURPOSE, OR ERROR-FREE.
5.2 ESTIMATES AND RECOMMENDATIONS ONLY. AI OUTPUTS ARE ESTIMATES, PROJECTIONS, RECOMMENDATIONS, SCORES, CLASSIFICATIONS, OR INFORMATIONAL NARRATIVES. THEY ARE NOT STATEMENTS OF FACT, AUDITED FIGURES, ASSURANCES, OR GUARANTEES OF ANY FINANCIAL, OPERATIONAL, YIELD, PRICING, QUALITY, COMPLIANCE, OR OTHER OUTCOME. FORWARD-LOOKING AI OUTPUTS (SUCH AS REVENUE OR YIELD PROJECTIONS) ARE INHERENTLY UNCERTAIN AND ACTUAL RESULTS MAY DIFFER MATERIALLY.
5.3 NO WARRANTY. EXCEPT AS EXPRESSLY AND UNAMBIGUOUSLY STATED IN A SIGNED WRITING BY CANNVAS, CANNVAS DISCLAIMS ALL WARRANTIES WITH RESPECT TO AI OUTPUTS, WHETHER EXPRESS, IMPLIED, OR STATUTORY, INCLUDING ANY IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, ACCURACY, TITLE, AND NON-INFRINGEMENT, TO THE FULLEST EXTENT PERMITTED BY APPLICABLE LAW.
6.1 NOT PROFESSIONAL ADVICE. AI OUTPUTS ARE PROVIDED FOR GENERAL INFORMATIONAL AND DECISION-SUPPORT PURPOSES ONLY AND DO NOT CONSTITUTE, AND ARE NOT A SUBSTITUTE FOR, PROFESSIONAL ADVICE. AI OUTPUTS ARE NOT FINANCIAL, INVESTMENT, ACCOUNTING, TAX, LEGAL, REGULATORY, COMPLIANCE, HORTICULTURAL/AGRONOMIC, MEDICAL, OR HEALTH ADVICE, AND DO NOT CREATE ANY PROFESSIONAL, FIDUCIARY, ADVISORY, OR OTHER SPECIAL RELATIONSHIP BETWEEN CANNVAS AND ANY ORGANIZATION, AUTHORIZED USER, OR CONSUMER.
6.2 CONSULT QUALIFIED PROFESSIONALS. BEFORE ACTING ON ANY AI OUTPUT, YOU SHOULD CONSULT YOUR OWN QUALIFIED PROFESSIONALS (SUCH AS ACCOUNTANTS, ATTORNEYS, TAX ADVISORS, AND COMPLIANCE PERSONNEL) AND VERIFY REQUIREMENTS WITH THE APPLICABLE REGULATORY AUTHORITIES AND SYSTEMS OF RECORD.
6.3 Compliance is your responsibility. Cannvas is ancillary software that is designed to reflect, rather than author, your regulatory system-of-record state. CANNVAS DOES NOT DETERMINE, CERTIFY, OR GUARANTEE YOUR COMPLIANCE WITH ANY LAW, REGULATION, LICENSE CONDITION, OR TRACK-AND-TRACE OBLIGATION. THE AUTHORITATIVE RECORD FOR REGULATORY COMPLIANCE IS THE APPLICABLE GOVERNMENT/TRACK-AND-TRACE SYSTEM AND YOUR OWN BOOKS AND RECORDS, NOT ANY AI OUTPUT.
7.1 NO SOLE RELIANCE. YOU MUST NOT RELY ON ANY AI OUTPUT AS THE SOLE OR PRIMARY BASIS FOR ANY FINANCIAL, OPERATIONAL, PRICING, INVENTORY, HORTICULTURAL, COMPLIANCE, LEGAL, OR OTHER DECISION. AI OUTPUTS ARE INPUTS TO — NOT REPLACEMENTS FOR — YOUR OWN ANALYSIS AND JUDGMENT.
7.2 INDEPENDENT VERIFICATION REQUIRED. YOU ARE SOLELY RESPONSIBLE FOR INDEPENDENTLY REVIEWING, VALIDATING, AND VERIFYING EVERY AI OUTPUT — AGAINST YOUR SOURCE DATA, YOUR SYSTEMS OF RECORD, AND APPLICABLE LAW — BEFORE RELYING ON OR ACTING ON IT. YOU MUST EXERCISE YOUR OWN INDEPENDENT PROFESSIONAL JUDGMENT AND SKILL. IF AN AI OUTPUT CONFLICTS WITH YOUR SOURCE DATA, SYSTEM OF RECORD, OR APPLICABLE LAW, THE LATTER CONTROL.
7.3 ASSUMPTION OF RESPONSIBILITY. ANY DECISION YOU MAKE, AND ANY ACTION YOU TAKE OR FAIL TO TAKE, IN CONNECTION WITH AN AI OUTPUT IS MADE AT YOUR OWN DISCRETION AND RISK, AND YOU — NOT CANNVAS — ARE RESPONSIBLE FOR THE CONSEQUENCES.
8.1 DECISION SUPPORT, NOT DECISION-MAKING. CANNVAS IS DESIGNED AS A HUMAN-IN-THE-LOOP, DECISION-SUPPORT TOOL. IT FLAGS, ESTIMATES, AND RECOMMENDS; IT DOES NOT DECIDE FOR YOU. THE SERVICE DOES NOT AUTOMATICALLY EXECUTE, SUBMIT, OR APPLY OPERATIONAL, FINANCIAL, OR COMPLIANCE ACTIONS ON YOUR BEHALF BASED ON AI OUTPUTS; A HUMAN AUTHORIZED USER MUST REVIEW AND AFFIRMATIVELY DECIDE WHETHER, WHEN, AND HOW TO ACT.
8.2 Human accountability. The Organization and its Authorized Users remain the decision-makers and the accountable parties for all actions taken within or informed by the Service. Where the Service surfaces a recommendation, flag, or suggested action, it does so for human review only, and any resulting change is initiated and confirmed by a human.
8.3 No delegation of professional duty. Use of the Service does not transfer to Cannvas any duty of care, professional obligation, licensing obligation, or supervisory responsibility that you owe under applicable law or to third parties.
9.1 Our good-faith measures. Cannvas employs good-faith, best-effort measures intended to REDUCE (but which cannot ELIMINATE) the risk of error in AI Outputs, which may include, among others:
9.2 BEST-EFFORT, NOT A WARRANTY. THE MEASURES DESCRIBED IN SECTION 9.1 ARE GOOD-FAITH, BEST-EFFORT ENGINEERING PRACTICES. THEY ARE NOT A GUARANTEE, REPRESENTATION, WARRANTY, OR SERVICE-LEVEL COMMITMENT OF ACCURACY, COMPLETENESS, OR ANY OTHER OUTCOME, AND THEY DO NOT REDUCE, WAIVE, OR SHIFT YOUR DUTY TO INDEPENDENTLY VERIFY AI OUTPUTS UNDER SECTION 7. THE DESCRIPTION OF THESE MEASURES IS PROVIDED FOR TRANSPARENCY AND MUST NOT BE CONSTRUED AS A PROMISE THAT ANY PARTICULAR AI OUTPUT IS ACCURATE OR ERROR-FREE. Cannvas may modify, add, or discontinue any such measure at any time without notice.
10.1 In-product notices. In addition to this Disclaimer, the Service may present point-of-use notices, labels, badges, tooltips, or acknowledgments at or near the location where AI Outputs appear (for example, indicating that content is AI-generated or a recommendation/estimate, or that data is insufficient). These in-product notices supplement, and do not replace or narrow, this Disclaimer.
10.2 Layered disclosure. The absence of a point-of-use notice on any particular screen or AI Output does not mean the item is not an AI Output or is exempt from this Disclaimer. This Disclaimer applies to all AI Outputs regardless of whether a specific in-product notice is displayed.
10.3 ACCEPTANCE BY USE. BY ACCESSING OR USING ANY AI FEATURE OR AI OUTPUT, OR BY CLICKING, DISMISSING, OR PROCEEDING PAST ANY IN-PRODUCT AI NOTICE, YOU ACKNOWLEDGE THAT YOU HAVE READ, UNDERSTOOD, AND AGREE TO THIS DISCLAIMER.
11.1 Informational only. Certain Consumer Apps surface AI-derived descriptive content, such as aroma/terpene ("sensory") analysis, flavor or class descriptions, and similar informational content. THIS CONTENT IS PROVIDED FOR GENERAL INFORMATIONAL AND EDUCATIONAL PURPOSES ONLY, IS SUBJECTIVE AND ESTIMATED, AND MAY NOT REFLECT THE ACTUAL CHARACTERISTICS, COMPOSITION, OR EXPERIENCE OF ANY PARTICULAR PRODUCT OR BATCH.
11.2 NO MEDICAL OR HEALTH CLAIMS. AI-DERIVED CONSUMER CONTENT IS NOT MEDICAL, HEALTH, THERAPEUTIC, DIAGNOSTIC, OR TREATMENT ADVICE, AND MAKES NO CLAIM REGARDING SAFETY, EFFICACY, HEALTH BENEFITS, OR INTENDED EFFECTS. IT IS NOT INTENDED TO DIAGNOSE, TREAT, CURE, OR PREVENT ANY DISEASE OR CONDITION. THESE STATEMENTS HAVE NOT BEEN EVALUATED BY THE U.S. FOOD AND DRUG ADMINISTRATION OR ANY OTHER REGULATORY AUTHORITY.
11.3 Not lab results; verify on the label/COA. AI-DERIVED SENSORY CONTENT IS NOT A CERTIFICATE OF ANALYSIS, LABORATORY RESULT, OR REGULATED PRODUCT LABEL. CONSUMERS SHOULD RELY ON THE OFFICIAL PRODUCT LABEL, THE APPLICABLE CERTIFICATE OF ANALYSIS (COA), AND LICENSED-RETAILER AND MEDICAL-PROFESSIONAL GUIDANCE, NOT ON AI OUTPUTS.
11.4 Age, legality, and availability. Consumer Apps concern regulated cannabis products. Availability, legality, and product information vary by jurisdiction and over time and may be inaccurate or out of date. Nothing in the Consumer Apps is an offer to sell, or a solicitation to buy, cannabis in any jurisdiction where doing so is unlawful. Use is subject to the Consumer Terms and applicable age and eligibility requirements.
12.1 AI-interaction transparency. Where the Service interacts with a person in a conversational or generative manner, Cannvas seeks to make clear that the person is interacting with an AI/Automated System. This Disclaimer, together with in-product notices, is intended to support applicable transparency expectations.
12.2 Evolving legal frameworks. AI-specific transparency, labeling, and disclosure obligations are evolving (including, for awareness and where applicable, U.S. Federal Trade Commission expectations regarding truthful, substantiated AI-related claims, and the European Union Artificial Intelligence Act's transparency obligations under Article 50, with key transparency provisions scheduled to apply on August 2, 2026). Cannvas intends to adapt this Disclaimer and its in-product notices as such requirements are finalized and as applicable to the Service. THIS SECTION IS PROVIDED FOR AWARENESS ONLY AND IS NOT A REPRESENTATION OR WARRANTY OF COMPLIANCE WITH THE LAW OF ANY PARTICULAR JURISDICTION.
13.1 Governing limitations apply. THE LIMITATIONS OF LIABILITY, DISCLAIMERS, EXCLUSIONS OF DAMAGES, AND INDEMNIFICATION PROVISIONS SET FORTH IN THE MSA (FOR ORGANIZATIONS AND AUTHORIZED USERS) AND THE CONSUMER TERMS (FOR CONSUMERS) APPLY IN FULL TO THIS DISCLAIMER AND TO ALL AI OUTPUTS AND ARE INCORPORATED HERE BY REFERENCE.
13.2 No liability for reliance on AI Outputs. TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, CANNVAS AND ITS AFFILIATES, LICENSORS, AND SUPPLIERS WILL HAVE NO LIABILITY FOR ANY LOSS OR DAMAGE OF ANY KIND — INCLUDING LOST PROFITS OR REVENUE, LOST OR SPOILED PRODUCT OR YIELD, PRICING OR INVENTORY LOSSES, REGULATORY OR COMPLIANCE PENALTIES, OR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, EXEMPLARY, OR PUNITIVE DAMAGES — ARISING OUT OF OR RELATING TO ANY AI OUTPUT OR ANY DECISION MADE OR ACTION TAKEN OR NOT TAKEN IN RELIANCE ON ANY AI OUTPUT.
13.3 Allocation of risk. THE DISCLAIMERS AND LIMITATIONS IN THIS DISCLAIMER REFLECT A REASONABLE ALLOCATION OF RISK AND ARE AN ESSENTIAL BASIS OF THE BARGAIN BETWEEN THE PARTIES. THEY APPLY REGARDLESS OF THE FORM OR THEORY OF ACTION AND EVEN IF ANY REMEDY FAILS OF ITS ESSENTIAL PURPOSE. Some jurisdictions do not allow certain exclusions or limitations, so certain of the above may not apply to you; in that case such exclusions or limitations apply to the fullest extent permitted by applicable law, and nothing here excludes liability that cannot be excluded by law.
BY ACCESSING OR USING ANY AI FEATURE OR AI OUTPUT OF THE SERVICE, YOU ACKNOWLEDGE AND AGREE THAT: (A) AI/AUTOMATED SYSTEMS GENERATE THE AI OUTPUTS; (B) AI CAN — AND SOMETIMES WILL — MAKE MISTAKES, AND AI OUTPUTS MAY BE INACCURATE, INCOMPLETE, OUTDATED, OR WRONG; (C) AI OUTPUTS ARE ESTIMATES AND RECOMMENDATIONS FOR DECISION SUPPORT ONLY, ARE NOT PROFESSIONAL ADVICE, AND ARE NOT GUARANTEES OF ANY OUTCOME; (D) YOU WILL NOT RELY ON AI OUTPUTS AS THE SOLE BASIS FOR ANY DECISION AND WILL INDEPENDENTLY VERIFY THEM AND EXERCISE YOUR OWN JUDGMENT; (E) A HUMAN — NOT CANNVAS — MAKES AND IS RESPONSIBLE FOR THE ULTIMATE DECISION; AND (F) YOU ACCEPT THIS DISCLAIMER, INCLUDING ITS LIMITATIONS OF LIABILITY.
15.1 Cannvas may update this Disclaimer from time to time to reflect changes in the Service, our practices, or applicable law. When we make material changes, we will update the "Last updated" date above and, where required or appropriate, provide additional notice (which may be through the Service). Changes are effective as of the "Last updated" date unless otherwise stated. Your continued access to or use of any AI feature or AI Output after an update constitutes acceptance of the updated Disclaimer. We encourage you to review this Disclaimer periodically.
Questions about this Disclaimer may be directed to Cannvas at [email protected] (attn: Legal / Cannvas, LLC).
*The following notes are for internal/attorney use only and are not part of the published Disclaimer. They flag drafting assumptions and open items and identify the principal frameworks that informed this draft.*
1. Placeholders / integration. Confirm the legal entity name and d/b/a, the "Last updated"/effective date, and the defined-term set so that this Disclaimer's definitions match verbatim those in the MSA and Consumer Terms. Confirm the incorporation-by-reference mechanics and the conflicts/precedence order (this draft makes this Disclaimer control *as to AI Outputs* only).
2. Layered / point-of-use disclosure. Current best practice (and improved enforceability) favors disclosures that appear where users actually encounter AI Outputs, not solely in back-end legal terms. Sections 3.2, 10, and 14 assume the product team will maintain conspicuous in-product AI labels/acknowledgments; align this Disclaimer with the actual UI (confirm what notices exist, their wording, and whether any require an affirmative click-through vs. browsewrap). Consider a click-through acknowledgment for high-consequence AI Outputs.
3. FTC (U.S.) considerations. Under Section 5 of the FTC Act and recent FTC AI-related guidance/enforcement (e.g., "Operation AI Comply"), (a) marketing/UI claims about the AI's capabilities or accuracy must be truthful and substantiated, and (b) a disclaimer does not cure an unsubstantiated performance claim. Recommend a companion review of all *marketing* and *in-product* claims about the AI (e.g., "grounding," "accuracy," "non-hallucination") to ensure competent and reliable substantiation; keep documentation of testing/validation and known limitations/failure modes. Avoid overstating the Section 9 measures.
4. EU AI Act (awareness/applicability). If the Service is offered to, or its outputs are used by, persons in the EU, assess Article 50 transparency obligations (AI-interaction notice; machine-readable marking/detectability of AI-generated content; deepfake and certain public-interest text labeling), with key provisions applying August 2, 2026 (and a grace period to Dec. 2, 2026 for pre-existing systems' machine-readable marking). Determine provider vs. deployer roles and whether any Cannvas outputs constitute in-scope "AI-generated content" requiring marking. Also assess state-law AI/automated-decision and consumer-transparency requirements as applicable.
5. Cannabis-specific / consumer content. Confirm Section 11 aligns with cannabis advertising/labeling rules in each operating jurisdiction, health-claim restrictions, and age-gating/eligibility. Confirm the "not FDA evaluated / not intended to diagnose…" language is appropriate for the product categories surfaced and does not itself imply a health claim.
6. Liability / warranty conformity. Reconcile Sections 5.3 and 13 with the MSA/Consumer Terms warranty disclaimers, damages caps/exclusions, indemnities, insurance, and governing-law/venue provisions to avoid inconsistency; verify consumer-protection carve-outs for the Consumer Terms audience. Confirm capitalization/formatting of conspicuous terms satisfies UCC/"conspicuousness" standards in the governing jurisdiction.
7. Automated decision-making / privacy. If AI Outputs feed any decisions about individuals, assess GDPR Art. 22 / state privacy (profiling, ADMT) notice and opt-out obligations; cross-reference the Privacy Policy/DPA.