ANNEX C — AI TERMS

These AI Terms are an annex to, and an integral part of, the Enastro General Terms and Conditions (“Terms”) published at https://dev.enastro.com/legal/terms, and govern the supply, deployment and use of the artificial intelligence systems incorporated in the Service. Capitalised terms not defined here have the meanings given in the Terms. “You” and the “Customer” mean the person who accepted the Terms; “SmartRep”, “we” and “us” mean SMARTREP S.A..

We supply the AI System as a standardised, multi-tenant software-as-a-service offering. It is neither designed, presented nor supplied for any use classified as high-risk under the AI Act, and the allocation of obligations and risk set out below reflects that position and the Fees agreed under the Terms.

1. Definitions

1.1 In these AI Terms:

(a) “Affected Persons” means natural persons whose data is processed by, or who are otherwise exposed to the operation of, the AI System in the course of your use of the Service, including your employees, agents and end customers.

(b) “AI Act” means Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, (EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828. Any reference to Articles or Annexes shall mean the Articles or Annexes of the AI Act.

(c) “AI System” means the machine-based components of the Service which are designed to operate with varying levels of autonomy and which, for explicit or implicit objectives, infer from the input they receive how to generate Outputs, as described in the Appendix to these AI Terms, including the transcription, diarisation, conversation analysis, metric extraction and natural-language query components, together with any successor, replacement or additional component which we make available from time to time. Components of the Service which do not meet the definition in Article 3(1) of the AI Act, including rule-based automation, reporting and data-processing functionality, are not AI Systems.

(d) “Classification Guidelines” means the guidelines on the classification of high-risk AI systems issued by the European Commission pursuant to Article 6(5) of the AI Act, as published and updated from time to time, together with the Commission guidelines on the definition of an artificial intelligence system.

(e) “High-Risk AI System” means an AI system classified as high-risk under Article 6(1) or Article 6(2) of the AI Act, read together with Annexes I and III to it.

(f) “Input Data” means data provided to, or directly acquired by, the AI System on the basis of which it produces an Output.

(g) “Instructions for Use” means the documentation which we make available from time to time describing the Intended Purpose, capabilities, limitations, conditions of use and human oversight measures of the AI System, including the Appendix to these AI Terms and any product documentation, model cards or trust-centre materials referenced in it.

(h) “Intended Purpose” means the use for which the AI System is intended by us, including the specific context and conditions of use, as specified in the Appendix to these AI Terms, the Instructions for Use, our promotional and sales materials and statements and the technical documentation.

(i) “Substantial Modification” means a change to the AI System after its placing on the market or putting into service which is not foreseen or planned by us and as a result of which the compliance of the AI System with the requirements of the AI Act is affected, or which results in a modification of the Intended Purpose.

1.2 “Deployer” and “provider” have the meanings given to them in Article 3 of the AI Act. References to Articles and Annexes are to Articles of, and Annexes to, the AI Act.

2. Roles

2.1 In respect of the AI System, we act as the provider and you act as the deployer, using it under your own authority, in your own name and for your own business purposes.

2.2 Each Party is solely responsible for compliance with the obligations which the AI Act and other applicable law impose on it in its own role, and for the cost of that compliance. We are not responsible for your compliance with your obligations as deployer or Provider, for your use case, for the lawfulness of your underlying operations, or for any decision you take on the basis of an Output.

2.3 You are responsible for determining whether your own use of the AI System is lawful in your sector and jurisdiction, and for carrying out any assessment, notification, consultation, registration or authorisation which applicable law requires of you. We do not advise you on these matters, and any information we make available is supplied for information only and constitutes neither legal advice nor an assurance of your compliance.

3. Classification

3.1 We warrant that, as at the date of these AI Terms and by reference to the Intended Purpose: (a) the AI System does not constitute, and is not designed or supplied to enable, a practice prohibited under Article 5; (b) the AI System is not a High-Risk AI System, is not intended for any use case listed in Annex III, is not intended to be used as a safety component of a product covered by Annex I, is not itself such a product, and is not subject to a third-party conformity assessment under the Union harmonisation legislation listed in Annex I; and (c) the AI System is lawfully placed on the Union market and put into service.

3.2 We have carried out and documented the classification assessment for which we are responsible under Article 6, taking into account the Classification Guidelines, and review it at least annually and upon any Substantial Modification, any amendment of Annex III, any material change in the Classification Guidelines and any material change in the functionality of the AI System.

3.3 You acknowledge that the classification of an AI system under the AI Act is determined by its intended purpose, and that a use which is not in accordance with the Intended Purpose constitutes reasonably foreseeable misuse. We give no warranty, and accept no liability, in respect of any classification arising from your own use case, configuration, integration, combination with other systems or deployment context.

3.4 We describe the Intended Purpose and the uses excluded under section 5 clearly, concretely and coherently across the Instructions for Use, the Terms, the Acceptable Use Policy, our promotional and sales materials and the technical documentation, and do not present, position or promote the AI System as suitable for any such excluded use. A contractual exclusion of high-risk uses is not by itself sufficient to prevent an AI system from being classified as high-risk where the provider’s overall presentation, examples or product positioning effectively provides for or promotes such uses.

3.5 You shall not describe, market, present, resell, sublicense or otherwise make available the Service or any Output to any third party in a manner which presents it as suitable for, or which promotes its use for, any use excluded under section 5.

3.6 You shall not configure or use the AI System, and shall not permit it to be used, for the profiling of natural persons within the meaning of Article 4(4) GDPR. An AI system falling within an area listed in Annex III which performs profiling of natural persons is in every case classified as high-risk and no exception under Article 6(3) is available for it.

3.7 The AI System is not supplied in reliance on any exception under Article 6(3), and we are accordingly not required to document an assessment under Article 6(4) or to register the AI System under Article 49(2). Deployment of the AI System in an area listed in Annex III in reliance on Article 6(3) is not permitted under the Terms.

3.8 You remain responsible for your own assessment of your use of the AI System and may address questions of classification to the AI Act Service Desk or to the competent market surveillance authority. The Classification Guidelines are not binding and any authoritative interpretation of the AI Act may ultimately be given only by the Court of Justice of the European Union.

3.9 If, as a result of an amendment of Annex III by delegated act under Article 7, of the annual review under Article 112(1), of a change in applicable law or in the Classification Guidelines, or of a change in the AI System, we determine that the AI System or any component of it falls or is likely to fall within the high-risk classification, we may modify the AI System or the Intended Purpose so as to remove that classification, or withdraw or discontinue the affected component or the AI System in accordance with Article 18 of the Terms. No obligation applicable to a provider or deployer of a High-Risk AI System is included in the Service or in the Fees.

3.10 Following Regulation (EU) 2026/1744, the requirements and obligations for High-Risk AI Systems classified under Article 6(2) apply from 2 December 2027 and those classified under Article 6(1) from 2 August 2028. Nothing in this paragraph obliges us to supply a High-Risk AI System.

4. Role change

4.1 Under Article 25(1) you will be considered to be the provider of a High-Risk AI System, and will assume all of the obligations of a provider under the AI Act, where you: (a) put your name or trade mark on the AI System; (b) make a Substantial Modification to it in such a way that it remains a High-Risk AI System; or (c) modify its intended purpose, including that of any general-purpose AI system made available through it, in such a way that it becomes a High-Risk AI System, including by any use in breach of section 5.

4.2 Where paragraph 4.1 applies: (a) we are discharged from any obligation under the AI Act arising from that change of role and cease to be considered the provider of the AI system concerned; (b) you shall not represent to any third party or competent authority that we are the provider of the resulting system; (c) our obligation to cooperate is limited to providing, at your cost and at our then-current rates, the technical information and assistance reasonably necessary to enable you to fulfil your obligations, subject to section 7.3; and (d) you shall indemnify us in accordance with clause 16.1 of the Terms.

4.3 You shall notify us in writing without undue delay, and in any event within five (5) Business Days, if you become aware of any circumstance falling within paragraph 4.1 or of any regulatory contact concerning the AI System.

5. Permitted and excluded uses

5.1 You shall use the AI System solely in accordance with the Intended Purpose and the Instructions for Use, within the scope of the Licence, in compliance with the Acceptable Use Policy and in compliance with applicable law. You shall ensure that every Member is informed of, is trained in, and complies with, the Instructions for Use, the restrictions in this section and the human oversight requirements in section 6.

5.2 You shall not use, and shall not permit any person to use, the AI System or any Output:

(a) for any practice prohibited under Article 5, including the use of an AI system to infer emotions of a natural person in the areas of the workplace or education institutions, save where the use is strictly for medical or safety reasons;

(b) as the sole or the determining basis for any decision concerning a natural person which produces legal effects concerning that person or similarly significantly affects that person within the meaning of Article 22 GDPR;

(c) for the recruitment or selection of natural persons, for decisions affecting the terms of work-related relationships, for the promotion, transfer, sanction or termination of work-related relationships, for the allocation of tasks based on individual behaviour or personal traits, or for the individual evaluation, ranking, scoring or disciplinary monitoring of natural persons in work-related relationships;

(d) to evaluate the eligibility of natural persons for essential private or public services or benefits, to evaluate creditworthiness or establish credit scores, to assess risk or pricing in life or health insurance, to evaluate or classify emergency calls, or for any other use case listed in Annex III;

(e) for any use in the areas of biometrics, critical infrastructure, education and vocational training, employment and workers management, law enforcement, migration, asylum and border control management, or the administration of justice and democratic processes;

(f) as a safety component of a product covered by Annex I, or in any product required to undergo a third-party conformity assessment under the Union harmonisation legislation listed in Annex I;

(g) to make the AI System available to any third party as a stand-alone artificial intelligence capability, to develop, train, fine-tune, benchmark or improve any artificial intelligence model or system which competes with the AI System, or to extract, distil or reconstruct any model, weights, prompt or architecture forming part of the AI System; or

(h) in any manner which would cause the AI System to be classified as a High-Risk AI System, or which would require us to comply with obligations applicable to providers of High-Risk AI Systems.

5.3 The AI System may produce Outputs concerning the conduct of your employees and agents in recorded conversations. You shall use such Outputs exclusively for aggregate quality assurance, process improvement, training-needs analysis and compliance monitoring at the level of your operations, and shall implement the safeguards required by section 6 and by employment, collective-bargaining and data-protection law before any Output is used in relation to an identified or identifiable individual.

5.4 Where we make available any functionality capable of inferring emotions or sentiment, you shall configure and use it only in respect of your own end customers and shall disable or exclude it in respect of your employees and other persons in work-related relationships. We make the means of doing so available in the Service. Failure to do so is a material breach of the Terms.

5.5 The restrictions in this section are conditions of the Licence. We may, without liability, suspend the affected functionality or the Service in accordance with clause 17.2 of the Terms where we reasonably believe that they have been breached.

6. Outputs and human oversight

6.1 You acknowledge and accept that the AI System is probabilistic and that, notwithstanding the exercise of reasonable skill and care by us: (a) Outputs are generated by statistical inference and may be inaccurate, incomplete, outdated, non-deterministic, internally inconsistent or factually wrong, and may present incorrect information as though it were correct; (b) identical or similar Input Data may produce different Outputs on different occasions, and a change of model or model version may change the Outputs; (c) transcription, speaker separation, language identification, sentiment and metric extraction accuracy vary with audio quality, background noise, language, dialect, accent, terminology, speech overlap and other factors outside our control; (d) confidence scores are indicative only and do not constitute a representation as to the accuracy of an Output; and (e) the AI System is not a substitute for professional judgement and does not provide legal, regulatory, financial, medical, employment or other professional advice.

6.2 You are solely responsible for evaluating the accuracy, completeness, suitability and lawfulness of any Output before relying on it, and for all decisions, actions and omissions based on or informed by an Output. You shall not present any Output to a third party as verified, as human-generated or as a statement of SmartRep.

6.3 We design and make available the AI System with human-machine interface features enabling you to exercise effective oversight proportionate to the risks of a non-high-risk AI system, including access to the underlying transcript and source recording supporting each Output, the ability to review, correct, override, annotate and delete Outputs, the ability to select, restrict or disable models, Jobs and functionality, and the ability to stop the processing of any Job.

6.4 You shall: (a) assign oversight of the AI System to natural persons who have the necessary competence, training and authority; (b) apply meaningful human review to any Output before it is used in relation to an identified or identifiable natural person, or in a manner producing legal or similarly significant effects; (c) remain aware of the possibility of automation bias; and (d) monitor the operation of the AI System in your own deployment context and intervene, or cease use, where an Output appears anomalous or inappropriate. We are not responsible for your exercise of oversight or for any consequence of your failure to exercise it.

7. Transparency, AI literacy and documentation

7.1 Where the AI System is intended to interact directly with natural persons, we design it so that the natural persons concerned can be informed that they are interacting with an artificial intelligence system, and make the necessary configuration available to you. As deployer, you are responsible for activating, maintaining and presenting that information in a clear and distinguishable manner, at the latest at the time of the first interaction, and in a form meeting applicable accessibility requirements.

7.2 Where the AI System generates synthetic audio, image, video or text content, we mark the Outputs in a machine-readable format and make them detectable as artificially generated or manipulated, to the extent technically feasible, effective, interoperable, robust and reliable, taking into account the state of the art, the costs of implementation and the timetable set out in the AI Act for the application of Article 50(2). We do not supply the AI System for the generation or manipulation of deep-fake content and you shall not use it for that purpose.

7.3 You are responsible, in your own capacity, for informing Affected Persons, including your employees and your end customers, that conversations are recorded, transcribed and analysed by an artificial intelligence system, and for the content, timing and legal sufficiency of that information; for any disclosure required under Article 50(4) where you publish text generated or manipulated by the AI System; and for consulting and informing employee representatives where required by applicable law.

7.4 Each of us shall take measures to ensure a sufficient level of AI literacy among its own staff and other persons dealing with the operation and use of the AI System on its behalf, in accordance with Article 4. We make available standard product documentation, release notes and self-service training materials for that purpose; bespoke training is a chargeable professional service.

7.5 We make available to you Instructions for Use containing clear, concrete and comprehensible information on the Intended Purpose, the functionalities, capabilities and known limitations of the AI System, the categories of Input Data it accepts, the human oversight features available and the uses for which it may not be used. Nothing in these AI Terms obliges us to disclose source code, model weights or parameters, system or user prompts, architecture, training methodologies, training, validation or testing datasets, evaluation results, security testing results, the terms of our agreements with Model Providers, or any other information which constitutes a trade secret or the disclosure of which would compromise the security of the Service. Any documentation disclosed is our Confidential Information, is provided for your internal compliance purposes only, and may not be published, benchmarked or disclosed to any competitor of ours.

8. Model Providers

8.1 The AI System routes inference requests to Model Providers through our internal gateway, and our ability to control the behaviour, availability, accuracy and roadmap of a third-party model is limited.

8.2 We maintain a current list of Model Providers, notify you of the addition or replacement of a Model Provider in accordance with the Data Processing Agreement, and engage each Model Provider on terms which prohibit the use of Customer Data for the training or improvement of that Model Provider’s own models, to the extent such terms are offered by the Model Provider concerned for the service in question.

8.3 We do not warrant that inference takes place within the European Economic Area or in any particular jurisdiction, and make no representation as to the residency of processing by Model Providers. Transfers of personal data are governed exclusively by the Data Processing Agreement.

8.4 We may add, substitute, deprecate or withdraw Model Providers, models and model versions in accordance with section 12. Where a Model Provider ceases to make a model available, or changes its terms in a manner which we reasonably consider unacceptable, we may substitute a functionally comparable model without your consent.

9. Customer Data, training and use restrictions

9.1 As between us, you retain all right, title and interest in the Customer Data. The licence granted to us in respect of the Customer Data is that set out in clause 9.3 of the Terms.

9.2 You warrant, represent and undertake on a continuing basis that: (a) you have obtained all recordings, transcripts and other Customer Data lawfully, and that all notices have been given, all consents obtained, all legal bases established and documented, and all obligations to your employees, their representatives and your end customers satisfied, as required for the recording, storage, transcription, analysis and other processing of Customer Data by us and our sub-processors; (b) our use of the Customer Data in accordance with the Terms does not and will not infringe any intellectual property right, trade secret, contractual restriction, right of confidence or right of any third party; (c) you will not upload any content which you are prohibited from disclosing to us, and will not deliberately or systematically introduce special categories of personal data within the meaning of Article 9 GDPR, or data relating to criminal convictions and offences, beyond what may incidentally arise in customer service conversations; (d) you have conducted any data protection impact assessment, legitimate interests assessment or consultation required of you; and (e) you will not direct the Service at children or deliberately use it to process conversations with persons known to you to be children.

9.3 We do not advise on, and assume no responsibility for, the lawfulness of your recording practices, your employee monitoring, your notices or your legal bases, and are entitled to rely on your warranties without further enquiry.

9.4 No training. We do not use the Customer Data to train, fine-tune, adapt, index, calibrate, evaluate or improve any model, AI system, product or service, whether for your benefit or for the benefit of any other customer or for our own general product development. Training on Customer Data is not available on the Free or Standard plan and may be agreed only in a separate written agreement signed by both of us.

9.5 We shall not, and shall procure that our personnel, contractors and sub-processors do not: (a) disclose Customer Data to any third party, except to sub-processors engaged in accordance with the Data Processing Agreement for the purpose of providing the Service, or as required by applicable law; (b) sell, licence, rent or otherwise make Customer Data available to any third party, or use Customer Data for advertising, profiling or benchmarking against other customers; (c) transmit Customer Data to a Model Provider otherwise than for the purpose of serving your own inference requests; or (d) access Customer Data except as necessary to provide, secure, support and maintain the Service, or as instructed by you.

9.6 Customer Data, your Confidential Information and your trade secrets remain confidential in accordance with Article 13 of the Terms. These obligations survive termination.

9.7 We may collect, generate and use Service Data and aggregated or anonymised data in accordance with clause 9.4 of the Terms. Nothing in these AI Terms restricts our right to use the general knowledge, skills, know-how, techniques and experience retained in the unaided memory of our personnel, provided that this does not entitle us to use or disclose Customer Data or your Confidential Information or trade secrets.

9.8 In respect of the data we use for the development and training of our own components of the AI System, we ensure that such data has been obtained and is used in a manner which does not infringe the intellectual property rights, trade secrets, database rights or contractual rights of third parties and, where personal data is involved, complies with Articles 6 and, where applicable, 9 GDPR. This paragraph does not apply to, and we give no warranty in respect of, the datasets used by Model Providers to train their own models.

10. Personal data, security and logs

10.1 The processing of personal data in connection with the AI System is governed exclusively by the Data Processing Agreement. You are the controller and we are the processor in respect of the Customer Data, save in respect of Service Data, aggregated data, account and billing data and security records, in respect of which we act as controller for our own purposes.

10.2 You are responsible for responding to requests from data subjects, for your own transparency obligations, for carrying out any data protection impact assessment and legitimate interests assessment required in respect of your use of the Service, and for any prior consultation with a supervisory authority.

10.3 We implement and maintain the technical and organisational measures set out in the Schedule to the Data Processing Agreement and operate the AI System so that it performs consistently under normal conditions of use and is resilient against errors, faults and inconsistencies, to a standard consistent with good industry practice for a non-high-risk AI system of this type. We take reasonable measures to protect the AI System against attempts by unauthorised third parties to alter its use, Outputs or performance, including measures addressing prompt injection, data poisoning, model evasion and adversarial input, to the extent technically feasible.

10.4 You shall protect your credentials, API Keys and integrations, shall not attempt to circumvent, disable or defeat any safety, security, filtering or rate-limiting control, shall not submit Input Data intended to cause the AI System to produce unlawful, harmful or infringing Outputs, and shall notify us without undue delay of any suspected compromise.

10.5 The AI System automatically records events relating to its operation, including job execution, model and version used, Input Data references, Outputs, corrections and access events. We retain such logs for the period stated in the Instructions for Use or the Data Processing Agreement and make available to you, through the Service, the logs relating to your own Organization. We are not obliged to provide access to logs relating to our internal infrastructure, our other customers, our Model Providers or our security operations, which constitute our Confidential Information. You are responsible for retaining logs and Outputs which you require for your own compliance, evidential or audit purposes, and for exporting them before termination.

11. Monitoring, incidents and regulatory cooperation

11.1 We monitor the operation of the AI System in respect of its conformity with the Intended Purpose, its resilience to errors and inconsistencies, and its security, at the level of the Service as a whole. We are not obliged to monitor your individual Outputs or use case.

11.2 You shall notify us without undue delay, and in any event within seventy-two (72) hours, of becoming aware of: (a) any malfunction of the AI System; (b) any Output which is, or is alleged to be, unlawful, discriminatory, defamatory, infringing or harmful; (c) any complaint, claim or regulatory contact relating to the AI System; or (d) any circumstance indicating that the AI System presents a risk within the meaning of the AI Act. You shall provide the information reasonably necessary for us to investigate and shall cease the affected use if we so require.

11.3 Where we establish that the AI System does not conform with these AI Terms, we shall take the corrective measures we consider appropriate and inform you of the non-conformity and of the measures taken. Those measures, together with the remedies in clause 14.2 of the Terms, are your sole and exclusive remedies in respect of non-conformity of the AI System.

11.4 Each of us shall notify the other without undue delay of any request, inquiry, investigation or order from a competent authority relating to the AI System, unless prohibited by applicable law. We lead any engagement with a market surveillance authority, the AI Office or any other competent authority in relation to the AI System as such, including in relation to its classification; you lead any engagement relating to your own deployment, use case or processing of personal data. We provide the information and assistance reasonably necessary for you to respond to a competent authority in relation to the AI System, subject to section 7.5; assistance exceeding four (4) hours in any twelve (12) month period, or required as a result of your breach of these AI Terms, is chargeable at our then-current rates. Neither of us shall represent to any authority or third party that the other has assumed a role or obligation under the AI Act which it has not assumed under these AI Terms.

12. Updates

12.1 We may at any time update, retrain, reconfigure, replace, upgrade, deprecate or withdraw the AI System, any model, model version or component, and may modify the Instructions for Use accordingly. Our model catalogue changes over time.

12.2 We give at least thirty (30) calendar days’ prior notice of any change which we reasonably expect to have a material adverse effect on the functionality of the AI System as you use it, or of the deprecation of a model you are using, save where the change is required to address a security vulnerability, a legal or regulatory requirement, a third-party licence restriction or the withdrawal of a model by a Model Provider, in which case we give such notice as is reasonably practicable.

12.3 We use reasonable endeavours to ensure that updates do not materially reduce the functionality or performance of the AI System, but give no warranty that Outputs will remain identical, consistent or comparable across models or versions. Variation in Outputs following a permitted change does not constitute a defect. We are not obliged to maintain or restore any previous version of a model or of the AI System, save that where we offer a model selection capability you may select among the models then available in the catalogue.

12.4 You shall not make, and shall not permit any third party to make, any modification to the AI System, and shall not integrate it with any other system in a manner which alters its Intended Purpose without our prior written consent.

13. Intellectual property, warranties and liability

13.1 We and our licensors retain all right, title and interest in and to the AI System, including all software, models, weights, adapters, prompts, architecture, documentation, Service Data and all improvements, modifications and derivative works of any of them. As between us, you own the Customer Data and, subject to our underlying rights in the Service, the Outputs generated for you, in accordance with clause 9.2 of the Terms. Outputs are generated by statistical inference, we may generate identical or substantially similar Outputs for other customers from other inputs, and we give no warranty that any Output is original, novel, unique or protectable.

13.2 We warrant that we will supply the AI System with reasonable skill and care and in accordance with good industry practice, that it will operate in material conformity with the Instructions for Use, and that the warranties in section 3.1 are true as at the date of these AI Terms. That warranty is given by reference to the AI System as a whole and not to any individual Output, and does not apply to the Free plan or to beta functionality.

13.3 To the maximum extent permitted by applicable law, and save as expressly stated in paragraph 13.2, the AI System and all Outputs are provided “as is” and we exclude all other warranties, conditions, representations and terms, whether express or implied, statutory or otherwise, including any warranty as to the accuracy, completeness, reliability, currency, originality, non-infringement or fitness for any particular purpose of any Output, the achievement of any particular result, level of accuracy, error rate, recall, precision or business outcome, the absence of bias, error, inconsistency or unexpected behaviour, the uninterrupted or error-free operation of the AI System, the availability, continuity or performance of any Model Provider, or the location of inference.

13.4 Our liability arising out of or in connection with these AI Terms, the AI System and any Output is subject to, and forms part of and is not additional to, the limitations and exclusions in Article 15 of the Terms. No presumption of causation, fault, defect or non-conformity arises against us by reason of the AI System having produced, or having failed to produce, an Output. Nothing in these AI Terms excludes or limits liability which cannot lawfully be excluded or limited, including under Directive (EU) 2024/2853 on liability for defective products.

13.5 You shall indemnify us in accordance with clause 16.1 of the Terms in respect of any breach of these AI Terms, any use excluded under section 5, and any circumstance falling within section 4.

14. General

14.1 In the event of conflict, the order of precedence in clause 2.4 of the Terms applies, under which the Data Processing Agreement prevails in respect of the processing of personal data and these AI Terms prevail, in respect of the AI System and the Outputs, over the Terms and over every other annex.

14.2 We may update these AI Terms in accordance with Article 18 of the Terms.

14.3 Sections 1, 5.5, 6.2, 9.6, 10, 13 and 14 survive termination, as does any provision which by its nature is intended to survive.

APPENDIX TO ANNEX C — DESCRIPTION OF THE AI SYSTEM AND INTENDED PURPOSE

Field	Content
Name and version of the AI System	Enastro, in the version made generally available by SmartRep at the Effective Date, as described in the Instructions for Use current at that date.
Components	Transcription and speaker separation; conversation analysis; Metric extraction; natural-language query over transcripts, in each case routed through the LLM Router
Intended Purpose	Analysis of recorded customer-service conversations for the purposes of quality assurance, process improvement, customer-experience measurement and business reporting at the level of the Customer’s operations
Context and conditions of use	Business-to-business use by the Customer’s customer-operations, quality-assurance and business functions, on conversations lawfully recorded by the Customer, with human review of Outputs
Categories of Input Data	Audio recordings of customer-service conversations; transcripts; call metadata; Job configuration
Categories of Output	Transcripts; speaker segments; summaries; topics; sentiment of end customers; action items; extracted Metric values with reasoning and confidence score
Excluded uses	The uses excluded under section 5, and in particular: employee evaluation, ranking, discipline, promotion or termination; emotion inference in respect of persons in work-related relationships; profiling of natural persons; automated decision-making with legal or similarly significant effects; any use case listed in Annex III; any use as a safety component under Annex I
Risk classification	Not a High-Risk AI System within the meaning of Article 6, assessed by reference to the Intended Purpose and taking into account the Classification Guidelines
Reasonably foreseeable misuse	Use of individual Outputs as a basis for decisions concerning identified employees; reliance on Outputs without human review; use in a regulated decision-making process; profiling of natural persons
Human oversight measures available	Access to source recording and transcript; review, correction, annotation and deletion of Outputs; confidence scores; model selection and restriction; job suspension and deletion
Transparency measures	Configurable notice that the system is artificial intelligence, where it interacts directly with natural persons; machine-readable marking of synthetic content
Model Providers	As set out in the published sub-processor and model list
Training on Customer Data	Not available on the Free or Standard plan