Lumana AI Policy

At Lumana, we believe artificial intelligence should be used responsibly, safely, and in ways that respect people’s rights. This policy sets out the conditions under which  our AI and services may be used by you, a Lumana customer. If you do not agree with these AI Terms please refrain from using the AI Functions.

These Artificial Intelligence Terms of Service  (the “AI Terms”) govern your access to and use of Lumana’s AI-powered products and tools (the “AI Functions”).  Lumana reserves the right to update the AI Terms from time to time without notice. You are encouraged to check these terms periodically. 

These AI Terms are incorporated by reference into Lumana’s Terms of Use available at https://www.lumana.ai/terms-of-use (the “Terms of Use”)

‍

1. AI Functions

1.1. Lumana’s AI Functions are integrated as part of Lumana’s Services, and they utilize and combine Artificial Intelligence, Large Language Models, image and video recognition and, if enabled, facial and other types of pattern recognition, with the aim of quickly and efficiently extracting analyzing data from video footage and still images. These capabilities can identify individuals and objects, and then analyze the output, thus replacing the need for manual human review, which is both time-consuming and less accurate.

1.2. The main AI Functions are carried out on the Core of the device, obviating the need to transfer any data back to Lumana’s Cloud Service. Lumana trains models on a per-camera basis, meaning that training is launched, and a dedicated model is pushed to that specific camera only. This process is fully automated and no human ever reviews the images or the labels. In certain instances, Lumana may also use data collected from the camera to train new models. This data is mixed together so that Lumana has no way of knowing where the data is coming from.

‍

2. AI Content

2.1. If you choose to use Lumana’s AI Functions, the AI Functions will receive image and video footage ("Input") and, following an analysis process will generate output based on an analysis of the Input ("Output", and together with Input, "Al Content").

2.2. By inputting the Input to the AI Functions, you acknowledge that you have all legal right and title to the Input, and you are permitted to input the Input into the AI Function without any restrictions. If any restrictions on your inputting of the Input exist, please refrain from using the AI Functions. Additionally, by using the AI Functions, you acknowledge and consent to Lumana sharing the AI Content with service providers as detailed in our Privacy Policy, in order to provide you with the AI Functions.

2.3. It is hereby clarified that you remain the owner of the AI Content.

2.4. Lumana exclusively owns all right, title, and interest in and to the Al Functions, including all associated intellectual property rights. No license or other rights (express or implied) in or to the AI Functions are granted, assigned, licensed, or conveyed to you.

‍

3. Your Consent and Acknowledgement

By using the Lumana AI Functions, you:

3.1. Expressly consent to the use of artificial intelligence (AI) features as part of Lumana’s Services. You acknowledge and agree that Lumana may collect, store, and process the AI Content, which may include personal data, as necessary to provide the AI Functions within Lumana’s Services. For the avoidance of doubt and as mentioned above, most of the AI Content is stored and processed on the Lumana Core which remains within your network. However, certain AI Content, such as minimized thumbnails may be sent to the Lumana Cloud for processing. In addition, AI Content may be accessed by Lumana engineers in the event that you request any support or assistance, or in the event that you engage Lumana to provide monitoring and alerting services on your behalf.

3.2. Represent and warrant that you have obtained all required consents and authorizations to allow Lumana to use the AI Functions for purposes contemplated hereunder, including notifying individuals of the use of the AI Functions, obtained their consent (to the extent legally required) before Input (including personal data) is used in the AI Functions, and you have secured all rights necessary to grant Lumana the ability to use the AI Functions in such manner. The Input will be used to provide the Services, to generate the Outputs, and for other purposes described in these AI Terms.

3.3. Acknowledge that your usage of the AI Functions and reliance on any Output is at your own risk. Lumana strongly recommends that you carefully evaluate the Output for accuracy and appropriateness for your specific use.

3.4. Are responsible for any AI Content and ensuring that it: (i) complies at all times with these AI Terms and the Terms of Use; (ii) does not violate any applicable laws and regulations of the jurisdictions which apply to you and your customers/users; (iii) does not infringe, violate, or misappropriate any of Lumana’s rights or the rights of any third party, including intellectual property rights; (iv) is free of any adverse claims and is not subject to any third-party rights, including using human review as appropriate, before using or sharing Output from the Services.

3.5. Acknowledge that Lumana may disclose such content to law enforcement or other governmental authorities, or in response to a court order, in the event that any of the AI Content is alleged to be unlawful or otherwise in breach of these AI Terms.

3.6. Give Lumana the right to host the AI Content on Lumana’s platform, comply with applicable laws, enforce Lumana’s policies and keep Lumana’s Services and users safe. It is clarified that hosting of the AI Content will remain local and correspond to the location of our customer.

‍

4. Restrictions

You hereby agree, when using the AI Functions, not to intentionally generate any harmful content, or to use any Output in a harmful way. 

All limitations which apply in the Terms and Conditions will apply to your usage of the AI Functions. In addition, it is prohibited to use the AI Functions to: (i) mislead anyone that the content generated by the AI Functions is human-generated; (ii) provide medical advice; (iii) provide legal or financial advice; (iv) generate spam, ransomware, viruses, intentional disinformation or deception, or other software, or to intentionally circumvent safety filters and functionality or prompt models to act in a manner that violates Lumana’s Terms of Use; (v) generate or disseminate information to be used for the administration of justice or other legal purposes; (vi) violate the privacy rights of others, including unlawful tracking, monitoring, and identification; (vii) depict a person’s voice or likeness without their consent or violate any other rights; (viii) harass, harm, or encourage the harm of individuals or specific groups; or to (ix) use the AI Functions for automated decision making that has legal or similarly significant effects on individuals, unless you do so in compliance with applicable law, and in such a case you acknowledge you are doing so at your own risk.

‍

5. Disclaimer of Warranties

Lumana is constantly working to improve its Services to make them more accurate, reliable, safe, and beneficial. Given the nature of artificial intelligence machine learning technology the AI Functions may, in some situations, result in an inaccurate Output. You acknowledge that Lumana has not verified the accuracy of the Output. Lumana makes no warranty or guarantee as to the accuracy,  completeness, fitness for a specific purpose, or reliability of the Output, nor the  legal possibility to use the Output or whether it infringes any third parties’ rights, and shall not have any liability or responsibility arising in any way regarding the Output or any omissions or errors contained in the Output. 

Lumana reserves the right to immediately, without reason and without notice, discontinue any or all AI Functions as deemed necessary, and you acknowledge and agree that you have no claims against Lumana should such action be taken. 

YOU ACKNOWLEDGE AND AGREE THAT LUMANA IS NOT RESPONSIBLE FOR ANY ACTIVITIES OR LEGAL CONSEQUENCES OF YOUR USE OF THE AI FUNCTIONS OR ITS OUTPUTS, INCLUDING IN CASE THAT A THIRD PARTY WILL HAVE A CLAIM AGAINST YOU FOLLOWING THE USAGE OF SUCH AI FUNCTIONS. 

THE AI FUNCTIONS ARE PROVIDED ON AN “AS IS” AND “AS AVAILABLE” BASIS. LUMANA HEREBY EXPRESSLY DISCLAIMS ALL WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, IMPLIED WARRANTIES OF MERCHANTABILITY, TITLE, NON-INFRINGEMENT AND FITNESS FOR A PARTICULAR PURPOSE, AS WELL AS ANY WARRANTIES REGARDING SECURITY, SUITABILITY, RELIABILITY, AVAILABILITY, TIMELINESS, ACCURACY AND PERFORMANCE OF THE AI FUNCTIONS. LUMANA MAKES NO WARRANTY THAT: (I) THE AI FUNCTIONS WILL MEET YOUR REQUIREMENTS, (II) THE AI FUNCTIONS WILL BE UNINTERRUPTED, TIMELY, SECURE, OR ERROR-FREE, (III) THE RESULTS AND/OR OUTPUT THAT MAY BE OBTAINED FROM THE USE OF THE AI FUNCTIONS WILL BE ACCURATE OR RELIABLE, OR (IV) ANY ERRORS IN THE AI FUNCTIONS WILL BE CORRECTED. THE FOREGOING DISCLAIMERS SHALL NOT APPLY TO THE EXTENT PROHIBITED BY APPLICABLE LAW. IN ADDITION, LUMANA DOES NOT WARRANT OR GUARANTEE THE ACCURACY OF THE RESULTS GENERATED BY THE AI FUNCTIONS OR THE AUTHENTICITY OF ANY IMAGES SUBMITTED. YOU REMAIN RESPONSIBLE FOR MAKING YOUR OWN BUSINESS DECISIONS AND FOR DETERMINING THE APPROPRIATENESS OF THE AI FUNCTIONS FOR YOUR UNIQUE USE CASE AND NEEDS.

‍

6. Limitation of Liability and Indemnity

IN NO EVENT SHALL LUMANA BE LIABLE, WHETHER IN TORT, CONTRACT, MISREPRESENTATION, RESTITUTION OR OTHERWISE, FOR ANY LOSS OF PROFITS, LOSS OF BUSINESS, DEPLETION OF GOOD-WILL AND/OR SIMILAR LOSSES, PURE ECONOMIC LOSS, OR FOR ANY SPECIAL, INDIRECT PUNITIVE, INCIDENTAL, CONSEQUENTIAL, EXEMPLARY OR OTHER SIMILAR LOSS, COSTS, DAMAGES, CHARGES OR EXPENSES, ARISING OUT OF OR IN ANY WAY CONNECTED WITH THE USE OF, MISUSE, INABILITY TO USE, OR THE RELIANCE UPON, THE AI FUNCTIONS AND/OR OTHERWISE IN CONNECTION WITH THESE AI TERMS AND/OR THE TERMS AND CONDITIONS; AND IN NO EVENT SHALL LUMANA’S TOTAL AGGREGATE LIABILITY, IN CONTRACT, TORT, MISREPRESENTATION, RESTITUTION OR OTHERWISE, ARISING OUT OF OR IN ANY WAY CONNECTED WITH THE USE OF, MISUSE, INABILITY TO USE, OR THE RELIANCE UPON, THE AI FUNCTIONS AND/OR OTHERWISE IN CONNECTION WITH THESE AI TERMS AND/OR THE TERMS AND CONDITIONS, EXCEED THE TOTAL FEES PAID BY YOU FOR THE AI FUNCTIONS DURING THE 12 MONTH PERIOD PRECEDING THE DATE IN WHICH THE CLAIM LEADING TO LIABILITY AROSE.

YOU AGREE TO DEFEND, INDEMNIFY AND HOLD LUMANA, AND ITS AFFILIATES, AND EACH OF ITS AND THEIR RESPECTIVE OFFICERS, DIRECTORS, AGENTS, CO-BRANDERS, OTHER PARTNERS, AND EMPLOYEES, HARMLESS FROM ANY AND ALL DAMAGE (WHETHER DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL OR OTHERWISE), LOSS, LIABILITY, COST AND EXPENSE (INCLUDING, WITHOUT LIMI-TATION, REASONABLE ATTORNEYS’ AND ACCOUNTING FEES) RESULTING FROM ANY CLAIM, DEMAND, SUIT, PROCEEDING (WHETHER BEFORE AN ARBITRATOR, COURT, MEDIATOR OR OTH-ERWISE), OR INVESTIGATION MADE BY ANY THIRD PARTY (EACH A “CLAIM“) DUE TO OR ARISING OUT OF YOUR USE OF, CONTRIBUTION TO OR CONNECTION WITH THE AI FUNCTIONS; YOUR VIOLATION OF THESE AI TERMS AND/OR YOUR VIOLATION OF ANY RIGHTS OF A THIRD PARTY INCLUDING VIOLATION OF PRIVACY OR INFRINGEMENT OF INTELLECTUAL PROPERTY RIGHTS OF A THIRD PARTY. LUMANA SHALL PROVIDE YOU NOTICE OF ANY SUCH CLAIM; PROVIDED, THAT, THE FAILURE OR DELAY BY LUMANA IN PROVIDING SUCH NOTICE SHALL NOT LIMIT YOUR OBLIGATIONS HEREUNDER. LUMANA RESERVES THE RIGHT TO ASSUME THE EXCLUSIVE DEFENSE AND CONTROL OF ANY MATTER WHICH IS SUBJECT TO INDEMNIFICATION UNDER THIS SECTION, AND IN SUCH CASE, YOU AGREE TO COOPERATE WITH ALL REASONABLE REQUESTS OF LUMANA IN ASSISTING LUMANA’S DEFENSE OF SUCH MATTER.

‍

7. Lumana’s Commitment

AI models are not perfect, they generate outputs based on patterns in data and may sometimes be inaccurate or inappropriate. It’s your responsibility to ensure that the Services are set up properly, ensuring that  outputs are reviewed and validated before being relied on.

You are also responsible for following all applicable laws, regulations and guidelines, such as the General Data Protection Regulation, as well as AI-specific requirements such as the EU AI Act.

Here are our commitments relating to the responsible development and deployment of our AI Functions:

• Industry Best-Practice: Lumana leverages proprietary datasets, foundational models and OpenAI’s moderation APIs, along with safeguards, to strengthen protection.

Development Phase

Requirement: Training data must be free from biases and reflect the diversity of society to avoid discrimination.

Lumana Compliance: We treat datasets as product specifications, curating data that is accurate, relevant, and representative. Where possible, we utilize publicly available datasets and validated open-source resources. We start with proven foundation models and enhance them with carefully selected data. We apply bias detection methods, diverse sampling, and quality checks to minimize discriminatory outcomes. For object detection Lumana fine tunes its models based on the customer’s scene and the distribution of objects in images at the customer’s site in order to increase the accuracy based on the customer’s dataset, all in good faith with no target of discrimination.

Deployment Phase

Requirement: The AI system must be safeguarded against tampering, and its environment must remain secure to preserve integrity.

Lumana Compliance: We implement strict security controls, encrypted storage, and access restrictions. Most of the raw footage remains within your environment, with Inputs and Outputs remaining local and never leaving the network. Our deployment pipelines are hardened against tampering, and environments are continuously tested to ensure robustness.

Operation Phase

Requirement: AI systems must be continuously monitored to remain ethical, fair, and aligned with intended use.

Lumana Compliance: We run ongoing automated monitoring, and periodic audits in order to test AI accuracy. Models are re-evaluated against updated datasets captured on customers' sites, and real-world scenarios to ensure fairness, transparency, and reliable performance over time. When new foundation models are adapted, we verify outputs continuously against both technical and ethical benchmarks. 

Lumana is committed to building AI that is safe, fair, and transparent. We provide guidance, and best practices to help you deploy and use AI responsibly. Together, we can ensure AI delivers positive impact while protecting people and communities.

‍

8. Removal Requests

To the extent that any AI Content is stored by Lumana, then if Lumana receives a removal request related to such AI Content (e.g. for infringing a third party’s intellectual property rights), Lumana may, at its sole discretion, taking into account technological limitation and the resources involved in carrying out such removal request, make its best efforts to remove such AI Content in whole or in part from the AI Functions, and will provide you with notice of the actions taken in response to such request. Lumana shall have no obligation to investigate such a request and shall have no liability in relation to its compliance, non-compliance, or partial compliance with any such request. Lumana reserves the right to notify the person or entity which made the removal request of such removal at its discretion. We welcome feedback on our AI Functions. If you have any questions or concerns, please do not hesitate to reach out to us at: [email protected].