AI Transparency

1. Why this page exists

Hi Networking uses artificial intelligence in several features. We publish this notice so you know, at a glance, what the AI does, which data it processes, who runs the model, and how you can opt out. It is a companion to our Privacy Policy.

2. AI features in Hi Networking

• Smart matching & connection suggestions — proposes people you may want to meet based on your stated interests, industry and role. Off unless you enable AI personalization.
• Profile completion assistant — suggests improvements to your bio and tags. Runs only when you click the assistant button.
• AI coach — answers your own networking questions. The coach never uses another user’s private data.
• Event recommendations — ranks public events for you.
• Conversation starters & summaries — produced on your device or in an EU-hosted model when you view a contact.
• Business-card OCR — extracts text from cards you scan. Images are processed transiently and not retained by the OCR provider.
• Voice-note transcription and translation — only for notes you record and only when you opt in for voice features.

3. AI providers we use

We route AI calls through the Lovable AI Gateway (Lovable GmbH, EU). The Gateway forwards requests to underlying models under a Data Processing Agreement that prohibits training on customer content. The current model set and their locations are listed on our Subprocessors page.

4. Data the AI sees

• Only the specific input required for the feature you triggered.
• Your identifiers are pseudonymised where possible before leaving our systems.
• Private messages are never sent to an AI provider unless you explicitly ask for a summary of your own conversation.
• Other members’ private data is never used to personalise your suggestions.

5. No training on your content

We do not, and our AI subprocessors do not, use your prompts, your uploads, your contacts, your messages or your business-card scans to train public foundation models. If we ever introduce optional opt-in improvement of Hi Networking’s own models, it will require a separate, explicit consent that you can withdraw at any time.

6. Automated decisions

Hi Networking does not make decisions that have legal or similarly significant effects on you purely by automated means. AI outputs are advisory: suggested matches, suggested summaries, ranked lists. You always decide whether to act on them. This aligns with your rights under Article 22 GDPR.

7. Your controls

• AI personalization toggle — Settings → Privacy Center. Turns off smart matching and personalised suggestions.
• AI features toggle — disables the AI coach, summaries and starters entirely.
• Per-feature opt-in — voice notes and business-card OCR are opt-in and shown with a clear notice before first use.
• Right to object — you can object to any AI-based processing under Article 21 GDPR by emailing privacy@hinetworking.com.

8. Retention

AI provider logs are retained by the provider for the shortest period compatible with abuse prevention (typically 30 days) and are then deleted. Outputs saved to your account (e.g. an accepted summary) follow the retention rules in our Privacy Policy.

9. Known limitations

Large language models can be wrong, biased or overconfident. Treat AI output as a starting point, not a source of truth. Do not paste other people’s confidential information into the coach. Report incorrect or harmful output using the flag icon on the response.

10. Contact

Questions about AI in Hi Networking: privacy@hinetworking.com.