AI Use Overview
Liana Technologies – Public Statement
v1.3 | September 2026
1. Our Commitment
Liana Technologies uses artificial intelligence in its Services and internal operations. We are committed to using AI responsibly, transparently, and in full compliance with the EU AI Act and GDPR. This document explains which AI systems we use in our products, how we classify them, and what commitments we make to customers regarding data, privacy, and human oversight.
This document is intended for customers, prospects, and the general public. It is published alongside Liana’s Data Processing Agreement and Subprocessors Register at lianatech.com/documents and reviewed at least annually.
Liana does not train AI models on Customer's Personal Data. This commitment covers personal data of End Users, Recipients, and any other data subjects processed through Liana's Services. It does not extend to non-personal content (such as newsletter text, press release, campaign copy, or other materials) that Customers or Users input into Liana's AI features, which Liana may use to improve its AI offerings unless that content contains personal data. This commitment applies to all AI features in all Liana Services and is legally binding under Liana's standard Data Processing Agreement.
This commitment applies to all AI features in all Liana services. It is legally binding under Liana’s standard Data Processing Agreement.
2. AI Features in Liana Services
The following AI features are currently available in Liana’s services:
| Service | AI Feature | What it Does | AI Provider | EU AI Act Risk Tier |
|
LianaMailer |
AI Writing Assistant |
Generates and improves email subject lines, preheaders, and body copy on request. All outputs require user approval before sending. |
Open AI / GPT models |
Limited Risk — Art.50 |
|
LianaMailer |
Newsletter Content Editor |
Suite of AI tools within the newsletter editor: generates subject line alternatives, body text, and images from user prompts. All outputs require user approval before sending. |
Open AI / GPT models |
Limited Risk — Art.50 |
|
LianaMailer |
Report Summaries |
Generates an AI summary of newsletter delivery report based on newsletter HTML content and aggregate statistics (no personal data such as recipient emails). |
Open AI / GPT models |
Limited Risk — Art.50 |
|
LianaMonitor |
AI Hits Summary |
Provides AI-generated summaries of media monitoring hits to help users quickly review relevant coverage. |
Google (Gemini)) |
Limited Risk — Art.50 |
|
LianaMonitor |
AI Hit Translation |
Translates monitoring hits into the user’s chosen language to support multilingual media monitoring workflows. |
Google (Gemini)) |
Limited Risk — Art.50 |
|
LianaAI Assistant |
Conversational AI assistant |
Assists marketing teams with content & image creation, campaign planning, audience segmentation, and send-time recommendations across Liana Cloud Services. |
Google (Gemini)) |
Limited Risk — Art.50 |
| LianaMonitor | Trending News | Provides the top trending news topics from online news, helping users quickly see what is being widely covered without manually reviewing individual articles. | Google (Gemini)) | Limited Risk — Art.50 |
| LianaAutoma tion |
AI Segmentation |
Lets users describe, in natural language, the kind of audience segment they want to create. The feature defines the segment's parameters based on that description for the user to review before use. |
Google (Gemini)) | Limited Risk — Art.50 |
| LianaMonitor | Smart Collection |
Filters media monitoring hits from selected monitors into a collection, based on a user-given textual instruction describing what should be included. | Google (Gemini)) | Minimal Risk — no Art.50 |
Limited Risk classification means these AI features interact directly with users and generatecontent, triggering transparency obligations under EU AI ActArt. 50. Users are informed theyare interacting with AI throughout the Service interface. One feature, LianaMonitor SmartCollection, filters existing content rather than generating new content or holding a direct conversational interaction with users; it is therefore classified as Minimal Risk and does not carry Art. 50 transparency obligations.
All AI-generated content requires explicit human approval before it is published, sent, or acted upon. No AI feature in Liana Services makes autonomous decisions with legal or similarly significant effects on individuals.
3. Data and Privacy
| Commitment | How we Deliver it |
|
No training on Personal Data |
All Liana AI features use enterprise APIs (Google Cloud Vertex AI and OpenAI) that contractually prohibit use of inputs for model training. Customer prompts and data are not retained by the AI provider beyond the API request. |
|
Personalisation based on user’s own preferences only |
LianaAI Assistant is personalised using preference settings stored per individual user (e.g. tone, style). No cross-customer profiling or data sharing. Users can update or delete their preferences at any time. |
|
GDPR compliance |
AI processing is documented in Liana’s Data Processing Agreements and DPIA for each affected product. Legal basis is Legitimate Interest (performance of service). Google Cloud and OpenAI are GDPR-compliant data processors. Full details are available in Liana’s Privacy Policy and DPA at lianatech.com/documents. |
|
Transparency in the Service interface |
Every AI-generated content suggestion is clearly labelled as AI-generated within the Service. Users are never in doubt about whether content was produced by a person or by AI. |
|
Human review of all AI outputs |
No AI output in any Liana Service is automatically published or sent. Every AI suggestion must be explicitly reviewed and confirmed by a human user before it takes effect. |
4. Internal AI Tools
In addition to the customer-facing AI features above, Liana staff use AI tools internally to support customer service and internal operations (for example, AI-assisted meeting transcription for internal communication summaries). These tools may process Customer Data such as contact details in the normal course of business.
Internal AI tools are not part of the services delivered to Customers and are therefore not listed in this overview. They are governed by Lianaʼs GDPR obligations and are listed as subprocessors in Lianaʼs Subprocessors Register at lianatech.com/documents. The same commitment applies to internal AI tools such as AI-assisted meeting transcription: Personal Data is never used for training. Non-personal content processed by these tools may be used to improve them, unless it contains personal data.
5. More Information
Customers and prospects can request the following documents at any time:
| Document | How to Obtain |
|
AI Systems Disclosure Register (full) |
Available to Customers prior to contract execution and on written request — privacy@lianatech.com |
|
Data Processing Agreement (DPA) |
Available at lianatech.com/documents |
|
Subprocessors Register |
Available at lianatech.com/documents |
|
Privacy Policy |
Available at lianatech.com/documents |
|
EU AI Act risk classifications (detailed rationale) |
Available on written request — privacy@lianatech.com |