EU AI Act - Customer Attestation

1.  Incorporation into the Agreement

This Customer Attestation (“Attestation”) is incorporated by reference into the Imagicle UCX Platform Cloud Service Agreement (“CSA”). By accepting the CSA — including through execution of an applicable Order Form that incorporates the CSA — the Customer acknowledges and accepts this Attestation as a binding part of the agreement between Imagicle and the Customer.

This Attestation sets out the Customer’s intended use of the Imagicle AI-enabled solutions and services and the minimum commitments adopted to support compliance with applicable obligations under Regulation (EU) 2024/1689 (the “EU AI Act”), as relevant to the Customer’s deployment context.

2.  AI Act Classification Reference

Imagicle’s classification of its AI-enabled products — Voice Analytics, AI Receptionist, Virtual Agent, and Advanced AI Analytics — under the EU AI Act, including the allocation of obligations between Imagicle (as provider) and the Customer (as deployer), is set out in the Imagicle AI Act Classification Statement, available EU AI Act - Classification Statement

Customers are encouraged to review the Classification Statement before deploying AI-enabled features. Upon request, Imagicle will also provide product-specific technical transparency documentation describing the AI components, data processed, and applicable safeguards.

3.  Customer Declarations

3.1  No decision-making about individuals in HR, credit, or health contexts

The Customer declares that the selected solutions/modules will NOT be used to monitor, evaluate, or make automated or semi-automated decisions about individuals in an employment context (e.g., performance evaluation, recruitment and selection, disciplinary actions), nor to make decisions in credit or healthcare contexts based on AI outputs.

3.2  Voice Analytics – specific limitations

If Voice Analytics is used, the Customer declares that:

• any insights and sentiment indicators are computed exclusively from transcribed text and do not reflect inference from biometric voice signals;

• no individual scoring or ranking of agents or users (“agent scoring”) will be generated or used for performance management purposes;

• the system will not be used to monitor or evaluate the performance or behaviour of persons in work-related contractual relationships, including employees, agents, or contractors.

3.3  Virtual Agent – specific limitations

If the Virtual Agent is used, the Customer declares that:

• the Virtual Agent will be deployed solely to support legitimate communication, information, and task-handling purposes, and will not employ subliminal, purposefully manipulative, or deceptive techniques that materially distort the behaviour of end users in a manner likely to cause significant harm (prohibited under Article 5(1)(a) of the EU AI Act);

• the Virtual Agent will not be used to exploit vulnerabilities of end users arising from their age, disability, or specific social or economic situation (prohibited under Article 5(1)(b) of the EU AI Act);

• end users will be clearly informed that they are interacting with an automated assistant, and the Virtual Agent will not be configured to impersonate a human or to conceal its artificial nature, consistent with Article 50 of the EU AI Act;

• the Virtual Agent will not be used to take automated decisions producing legal effects, or similarly significant effects, on individuals without meaningful human oversight.

3.4  AI Receptionist – specific limitations

If the AI Receptionist is used, the Customer declares that:

• the AI Receptionist will be used solely for call answering, routing, scheduling, and the provision of general information, and not to determine eligibility for, or access to, essential public or private services;

• callers will be informed that they are interacting with an automated assistant rather than a human operator, consistent with Article 50 of the EU AI Act;

• the AI Receptionist will not be used to perform biometric identification or biometric categorisation of callers, including any inference of sensitive characteristics from voice;

• any call routing or prioritisation will be based on operational criteria (such as topic, language, or availability) and not on automated profiling that produces detrimental effects on individuals.

3.5  Advanced AI Analytics – specific limitations

If Advanced AI Analytics is used, the Customer declares that:

• Advanced AI Analytics ingests Call Detail Records (CDR) and related telephony metadata from the calling platform solely to generate dashboards and reports on call management (such as call volumes, handling times, routing, and service levels), and not to process the content of communications or to make automated or semi-automated decisions about individuals;

• the dashboards and reports present aggregated operational and service-quality metrics, and will not be used to produce individual scoring, ranking, or profiling of agents or users for performance management, disciplinary, or other employment-related purposes;

• the analytics will not be used to monitor or evaluate the performance or behaviour of persons in work-related contractual relationships, including employees, agents, or contractors;

• where Advanced AI Analytics captures the costs associated with calls (including outgoing calls) and allocates them to individual users, such cost attribution will be used solely for telecom expense management, billing, and internal cost-accounting purposes, and not for performance evaluation, ranking, disciplinary action, or any automated or semi-automated decision affecting individuals;

• no call content, recordings, transcripts, biometric data, or emotion inference is processed by Advanced AI Analytics; the product operates exclusively on call metadata (CDR).

3.6  Transparency to end users – AI disclosure obligation

The Customer commits to informing end users when they are interacting with AI-enabled features — for example, through an audible statement at the start of a voice interaction or a clear notice at the start of a chat — and/or when calls may be transcribed and/or analysed using AI, as required by Article 50(1) of the EU AI Act and applicable law.

The Customer shall not disable, suppress, or obscure any AI interaction disclosure that Imagicle has built into the Products (e.g., the automated notification that the caller is speaking with a virtual assistant). Imagicle designs these disclosures to fulfil its obligations as provider under Article 50(1); interfering with them may constitute a breach of this Attestation and of the EU AI Act.

3.7  Privacy and GDPR responsibilities

The Customer, as data controller, remains responsible for determining the applicable legal basis for processing personal data (including any notice and consent requirements) and for carrying out required assessments (e.g., Data Protection Impact Assessment under Article 35 GDPR) in relation to the use of AI-enabled features, including call recording, transcription, and analytics.

3.8  Logging and auditability

The Customer agrees to enable and retain the technical and usage logs necessary to evidence compliance with the commitments in this Attestation and to support audits and compliance activities, with a minimum retention period consistent with applicable law and the Customer’s internal policies.

3.9  Incident reporting and cooperation

The Customer will promptly notify Imagicle of any significant incidents, malfunctions, or improper uses related to AI-enabled features and will cooperate with Imagicle on mitigation and continuous improvement actions.

4.  Prohibited Uses

The following uses are prohibited under this Attestation and under the EU AI Act. The Customer shall not use any Imagicle AI-enabled feature for:

• monitoring or evaluating the performance or behaviour of persons in work-related contractual relationships (including employees, agents, or contractors), including where AI outputs inform HR decisions;

• individual scoring or ranking of agents or users based on sentiment or conversation analytics for performance management purposes;

• emotion recognition or inference of emotions from biometric voice signals in employment or education contexts (prohibited under Article 5(1)(f) of the EU AI Act);

• automated credit or healthcare decisions based on AI outputs;

• the use of subliminal, purposefully manipulative, or deceptive techniques that materially distort a person’s behaviour in a manner that causes or is reasonably likely to cause significant harm (prohibited under Article 5(1)(a) of the EU AI Act);

• the exploitation of vulnerabilities of a person or group due to their age, disability, or specific social or economic situation in order to materially distort their behaviour (prohibited under Article 5(1)(b) of the EU AI Act);

• the evaluation or classification of individuals or groups over a period of time, based on their social behaviour or personal characteristics, leading to detrimental or unfavourable treatment (“social scoring”) (prohibited under Article 5(1)(c) of the EU AI Act);

• biometric categorisation of individuals to deduce or infer sensitive attributes such as race, political opinions, trade union membership, religious or philosophical beliefs, or sexual orientation (prohibited under Article 5(1)(g) of the EU AI Act);

• any other practice prohibited under Article 5 of the EU AI Act.

5.  Escalation – High-Risk Use Cases

This Attestation covers standard deployments of Imagicle AI-enabled features. If the Customer intends to deploy a Product for any use case listed in Annex III of the EU AI Act (including employment and worker management, access to essential services, or other high-risk contexts), this Attestation does not apply in its standard form and a dedicated compliance path will be required.

In such cases, the Customer shall contact Imagicle at trust@imagicle.com before proceeding with the deployment, to evaluate the applicable obligations and agree a suitable addendum. Imagicle reserves the right to withhold access to AI-enabled features for high-risk use cases until an appropriate written addendum has been executed.

6.  Legal Notice

This Attestation is provided in good faith based on the EU AI Act and the guidance available as of the effective date. It does not constitute legal advice and should not be relied upon as such. The EU AI Act is subject to ongoing implementation through Commission guidelines, codes of practice, harmonised standards, and pending amending legislation; this Attestation will be updated as appropriate.

Customers and partners remain responsible for assessing and meeting their own obligations as deployers under the EU AI Act, including any obligations arising from their specific use case that are not addressed in this Attestation.

In the event of any conflict between this Attestation and the terms of the CSA or the applicable Order Form, the order of precedence established in the CSA shall apply.