How Infragence AI handles information.

1. Information we may process

• Field-capture data — gang and material photos, headcount, GRN and progress reports submitted over WhatsApp, Telegram or web.
• Voice notes and their transcripts where lawfully enabled, plus the language and role of the sender.
• Project data — schedules, tasks, milestones, drawings, RFIs, submittals, punch lists and document versions.
• Financial and commercial data — budgets, actuals, contracts, change orders, RA bills, invoices, procurement records and tax details supplied by the customer.
• Quality, safety and compliance records — inspections, NCRs, incidents, permits and compliance-calendar items.
• Technical logs, workflow execution status, security events, API usage and service diagnostics.

2. How the workflow operates

Field submissions are normalised into a single project record. The AI may verify headcounts, OCR and three-way-match materials, draft daily progress reports, prepare RA bills, reorder notes and reports, score risk and forecast overruns, and surface leakage. Prepared financial or ERP entries are queued for a human to review and post. Each action may be written to an event log so owners can review source, actor and outcome.

3. Purposes and lawful use

We use information to provide the contracted service, run construction operations, secure and troubleshoot the platform, measure usage, improve reliability and comply with applicable law. Customers must provide appropriate privacy notices and obtain any required consent for messaging, voice capture, image processing and automated processing of their workers' and vendors' data.

4. Operating rules

• Do not fabricate quantities, rates, valuations, savings or compliance assurances.
• Use only project, financial and vendor information supplied or approved by the customer.
• Escalate unusual, sensitive, disputed or high-value decisions to an authorised human.
• Never post to a customer's ERP or accounting system without human review.
• Honour deletion, opt-out and access requests promptly.
• Restrict command-center access to authorised owners, managers and roles.

5. AI providers and service partners

To deliver the platform, Infragence AI may use vetted infrastructure, messaging, voice, analytics, automation and AI providers. Only the information needed for a task is shared. Providers process information under their own security and contractual commitments. We do not sell personal information.

6. Retention and deletion

Information is retained only as long as needed to provide the service, meet contractual or legal obligations, resolve disputes and maintain security. Customers may request deletion or export of eligible data, subject to lawful retention requirements and technical backup cycles.

7. Security and access

We use reasonable organisational and technical safeguards, including dedicated per-company environments, encryption, access controls, credential separation, event logging and least-privilege practices. Formal SOC 2 and ISO 27001 certifications are on our roadmap and not yet held. No digital service can guarantee absolute security; customers are responsible for protecting their credentials and promptly reporting suspected misuse.

8. Your choices and rights

Depending on applicable law (including GDPR and DPDP-style regimes), individuals may request access, correction, deletion, restriction or withdrawal of consent. Contact us with enough information to identify the relevant project and customer. We may need to verify identity before acting.

9. Contact

For privacy questions, rights requests or security concerns, contact Infragence AI: