Trust Centre

Welcome to Hailey's Trust Centre. Here, you'll find everything you need to know about our security, privacy and compliance practices.

AI should make everyday HR work easier, without compromising your data or your trust

That's why we've built our own AI system, held to the same standard as the rest of the platform: it has to be secure, stable and useful. This policy explains how our AI works, what data it uses and how we protect it. In short:

  • You're in control. AI is optional and turned off by default. Your admins decide which features to turn on, one module at a time.
  • Your data stays in the EU. Our AI runs on open models hosted by Scaleway, a European provider, on servers in France. Your data is never passed on to third-party AI services.
  • Your data is never used to train AI models.
  • People make the decisions. AI can suggest and draft, but your team always reviews and decides.

AI Policy  

1. Introduction 

This AI Policy explains how Hailey HR ("Hailey") integrates the optional use of artificial intelligence ("AI") within its services through its proprietary AI system, including AI-assisted feedback notes and templates, contract template generation, an organisational bot integrated with Teams and Slack, recruitment-related AI features, and AI-assisted survey templates. This policy serves as a transparency notice of the AI system. 

2. Definitions 

In this AI Policy, except were set forth otherwise, the following terms and abbreviations shall have the following meanings: 

“AI Act” refers to 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 (Artificial Intelligence Act); 

“AI system” means a machine-based system that is designed to operate with varying levels of autonomy and that exhibit adaptiveness after deployment, and that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations;  

“GDPR” means Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (General Data Protection Regulation, or “GDPR”) as well as supplementary local adaptions.; 

 “GPAI” or “General-purpose AI model” means an AI model, including where such an AI model is trained with a large amount of data using self-supervision at scale, that displays significant generality and is capable of competently performing a wide range of distinct tasks regardless of the way the model is placed on the market and that can be integrated into a variety of downstream systems or applications, except AI models that are used for research, development or prototyping activities before they are placed on the market; 

“Deployer” means a natural or legal person, public authority, agency or other body using an AI system under its authority where the AI system is used during a professional activity;  

“Provider” means a natural or legal person, public authority, agency or other body that develops an AI system or a general-purpose AI model or that has an AI system or a general-purpose AI model developed and places it on the market or puts the AI system into service under its own name or trademark, whether for payment or free of charge; 

3. Scope of AI-System  

Haileys AI system serves to assist users across multiple modules of the Hailey platform. AI capabilities currently include Generative AI and Contextual AI. The use of AI in Hailey is entirely optional. All AI settings are deactivated by default and must be explicitly activated on a per-module basis by admin users within the organisation, under Settings – Security and AI. 

AI functionality in Hailey 

Generative AI 

Generate and refine one-to-one notes, feedback templates, survey templates. 

Draft contract templates and generate creative and detailed job advertisements user-supplied information.  

Generate a concise summaries of Candidate CV’s including key education and work experience highlights. up to three bullet points per section, prioritised by recency. 

Contextual AI 

providing employees with access to in-depth organisational information, including preset information and contact information for colleagues such as work phone numbers and company emails. 

3. AI value chain 

Hailey uses a combination of open-source General-Purpose AI (GPAI) models, integrated into the AI system. AI inference is performed on European infrastructure operated by Scaleway S.A.S. A full overview of the value chain is available for customers upon request.  

3.1 Data Sources for AI Processing 

Depending on the AI feature modules the customer has enabled in their system environment, Hailey's AI features process data that may include: 

  • User-supplied prompts and inputs (notes, contract information, job advertisement briefs, survey descriptions), 
  • One-to-one feedback notes within the Feedback module,  
  • Jobb applications and CV’s,  
  • Organisational information accessible via the Bot module, including contact information for colleagues (work phone numbers and company emails) sourced through the customer's Teams or Slack integration,  
  • Conversations between end users and the AI system,  
  • Error and operational data captured through Hailey's monitoring infrastructure (Azure Application Insights). 

3.2 EU data residency 

AI inference for all AI models is performed by Scaleway S.A.S., a European service provider, ensuring that AI processing takes place within European infrastructure.  

3.3 Training on customer data 

No customer data will be used for training purposes.  

4. Roles under the AI Act 

Hailey is the downstream Provider of the AI system offered in Hailey within the meaning of Article 3(3) of the AI Act. Hailey integrates third-party GPAI models but is not the Provider of those underlying models, which remain the responsibility of the relevant model providers identified in section 4.2 above. Hailey's responsibility extends to the design, development, and offering of the AI functionality in compliance with applicable regulations. 

The Customer acts as the Deployer within the meaning of Article 3(4) of the AI Act when using Hailey's AI system during its professional activities.  

The Deployer are responsible for: 

  • Assessing whether its own use of the AI functionality results in a different classification under the AI Act; and 
  • Complying with any obligations resulting from such use, including the responsibilities set out in section 9 of this AI Policy.  

6. AI Act Classification 

Hailey’s AI system has been assessed as an AI system that falls under exemption set out in Article 6(3) Regulation (EU) 2024/1689 (the “AI Act”) and therefore is not considered a high-risk AI system. Hailey has documented this assessment and is provided per the customer’s request.  

7. Security 

7.1 Human Involvement 

Hailey's AI features are designed to assist users in their everyday task, but never to replace human judgment. 

  • All AI outputs, including refined notes, generated templates (feedback, contract, survey), drafted job advertisements, candidate assessment highlights, and bot responses, are solely intended as recommendations or assistive content. 
  • Users are responsible for reviewing all AI-generated content and no AI functionality shall be used for decision-making purposes. 
  • AI functionality is activated only by admin users on a per-module basis, ensuring that the deployment of AI is a deliberate and controlled decision made by each customer organisation. 
  • All AI-generated output must subject to human review and does not replace or influence prior human assessment without proper human review. 

7.2 Automated Decision-Making 

Hailey's AI features do not produce decisions that have legal or similarly significant effects on individuals within the meaning of Article 22 GDPR.  

8. Accuracy and Limitations 

AI outputs may contain errors, omissions, or misinterpretations. The AI system may produce incorrect or incomplete outputs across all modules, including but not limited to the Feedback module's note refinement functionality.  

  • AI features are provided on an “as is” basis. Hailey does not guarantee 100% accuracy of any AI output. 
  • In the Recruitment module, the end user is explicitly informed that the AI function is subject to limitations and that the tool must not be relied upon for decision-making purposes. 
  • All AI-generated outputs of the Service are clearly identified as AI-generated within the user interface, in accordance with Article 50 (1) of the AI Act. 

 9. Customer obligations & Fair use

Hailey’s AI features are provided as an integrated part of the Hailey services, for use by the Customer’s authorised users and within the intended purposes described in this AI Policy. To safeguard the availability, performance and security of the AI system, and to ensure fair access for all Customers, the following fair use rules apply.

9.1

The Customer shall ensure that its users comply with all instructions, guidance, limitations, and recommended practices provided by Hailey in relation to the use of AI features, including, but not limited to; (i) Ensuring that end users are adequately informed about how AI is utilized within Customer’s Hailey Account; (ii) ensuring that end users understand their obligation to review and validate all AI-generated content before use; (iii) ensuring that end users understand that AI-generated content must not be relied upon for decision-making purposes without appropriate human review and oversight.

9.2

AI features may only be used by the Customer’s authorised users, through the user interfaces and integrations provided or approved by Hailey, and for purposes connected with the Customer’s ordinary HR activities.

9.3

AI features shall not be used, configured or combined with other systems in a way that changes their intended purpose, including to make or determine decisions on recruitment, promotion, termination, allocation of tasks or the monitoring and evaluation of performance and behaviour of individuals.

9.4

Hailey’s AI features are designed and dimensioned for interactive use by individual human users. Hailey may apply reasonable usage limits to the AI features, such as limits on the number, frequency or size of requests per user or per Customer account, in order to protect the stability of the service for all customers. Such limits may be adjusted over time. Usage that materially exceeds what can reasonably be expected from normal human use of the features, such as continuous or high-frequency requests or inputs unrelated to the intended purpose, falls outside fair use. Unless expressly provided for by Hailey through an API or integration, or otherwise approved by Hailey in writing, the Customer and its users shall not;

  1. connect, deploy or allow autonomous AI agents, bots, scripts, browser automation, crawlers, third-party AI tools or other automated means to send prompts to, retrieve outputs from, or otherwise interact with Hailey’s AI features;
  2. share user credentials, or make the AI features available to third parties;
  3. use outputs to train, fine-tune, evaluate or benchmark other AI models; or,
  4. circumvent or attempt to circumvent usage limits, rate limits or other technical restrictions.

9.5

Hailey’s AI system includes safeguards, such as system instructions, content filters and refusals, which are an integral part of its design. A refusal by the AI system to respond is a deliberate safeguard. If a user considers a refusal to be incorrect, the user should report it through the error reporting function rather than attempt to work around it. The Customer and its users shall not:

  1. attempt to disable or manipulate these safeguards, including through so-called jailbreaking, prompt injection, role-play or hypothetical scenarios, encoded or obfuscated instructions, or repeated rephrasing intended to obtain output that the AI system has declined to provide;
  2. attempt to extract system instructions, configuration, model information, or data belonging to other users or Customers;
  3. use the AI features to generate content that is unlawful, harmful, discriminatory, harassing or defamatory, that facilitates violence, crime or self-harm, or that infringes the rights of others, including intellectual property and privacy rights;
  4. use the AI features for any practice prohibited under Article 5 of the AI Act, including to infer the emotions of natural persons in the workplace (Article 5(1)(f)), for social scoring (Article 5(1)(c)) or for manipulative or deceptive techniques (Article 5(1)(a));
  5. introduce malicious code, or conduct security or penetration testing of the AI system, without Hailey’s prior written consent.

9.6

Hailey may apply reasonable usage limits to the AI features, such as limits on the number, frequency or size of requests per user or per Customer account, to protect the stability of the service for all customers. Such limits may be adjusted over time. Usage that materially exceeds what can reasonably be expected from normal human use of the features, such as continuous or high-frequency requests or inputs unrelated to the intended purpose, falls outside fair use.

9.7

The Customer bears sole responsibility for all use of AI within its system environment, including the configuration, activation, and use of Hailey’s AI features by its users, and for complying with all applicable obligations of the Deployer under applicable law. The Customer further acknowledges and agrees that it is responsible for assessing the appropriateness of AI-generated outputs and for any decisions, actions, or outcomes arising from its or its users’ use of Hailey’s AI features.

10. Governance and lifecycle management 

10.1 Validation and change management 

Any changes to the AI system are subject to a structured validation process. When a change is initiated, Hailey performs regression testing using a comprehensive set of predefined conversations designed to verify that the system produces correct and expected outputs across all supported functionalities. The results are documented in a report that must be reviewed and approved by the designated Product Owner before the change is deployed to production. 

10.2 Data retention 

  • Conversations between end users and the AI system: retained in memory for a maximum of seven (7) days, after which they are automatically deleted. During the retention period, conversation data may be used for error identification and quality assurance purposes, provided that the end user choose to report an error. 
  • Error logs: Reported errors are stored by Hailey for further investigation for three (3) months from the date the error was reported. Access to the error log is restricted in accordance with the information security policy.  
  • AI-derived outputs saved by users: Storage depends on the customer’s specific retention policy.   

 Changes to This Policy 

This AI Policy may be updated at Haileys discretion to reflect technological, legal, or operational changes including changes to the underlying GPAI models or to the AI inference provider. 

Material updates will be communicated to customers.

Use of the services after publication constitutes acceptance of the updated policy. 

Latest edit: 2026-09-07

For questions about this AI Policy or Hailey's AI features, please contact:

The form could not be loaded because of your browser tracking settings. Contact us at hello@haileyhr.com.