verified_user
Singapore PDPA (2012) & Enterprise AI Governance Standard
Zero foundation model training on client data. Robust enterprise privacy protection tailored for Singapore and regional businesses.
01. Entity Scope & Applicable Law
ProjectAI LLP ("ProjectAI", "we", "us", or "our") is a partnership registered in the Republic of Singapore. This Privacy Policy governs the collection, use, processing, and protection of personal and corporate data in strict accordance with the Singapore Personal Data Protection Act 2012 (PDPA), advisory guidelines issued by the Personal Data Protection Commission (PDPC), and Singapore's Model AI Governance Framework.
02. Roles: Data Controller vs. Data Intermediary
To protect both ProjectAI and our enterprise clients, roles under the PDPA are defined as follows:
- Direct Website Inquiries: For inquiries submitted directly through this website or discovery forms, ProjectAI acts as a Data Controller / Organisation, collecting only necessary contact and project scoping details.
- Client AI Implementations: When deploying bespoke automations, private LLMs, or knowledge bases on client infrastructure, ProjectAI acts strictly as a Data Intermediary (Data Processor) under Section 4(2) of the PDPA. The Client remains the Data Controller responsible for securing all requisite employee, customer, and end-user consents.
03. Information We Collect & Lawful Processing Grounds
We collect and process information under the legal grounds of consent, legitimate business interest, and contractual necessity:
- Business Contact Information: Name, corporate email address, business WhatsApp/phone number, company designation, and entity name.
- Operational & Architecture Scoping: Industry domain, operational bottlenecks, software stack (e.g., CRM, ERP), and workflow parameters provided in consultation requests.
- Cybersecurity & Session Telemetry: Anonymized technical logs, browser types, and IP addresses strictly used for firewall security, anti-DDoS, and server auditability.
04. Zero Foundation Model Training Guarantee
shield
Enterprise Non-Training Covenant:
ProjectAI guarantees that client proprietary documents, prompt payloads, vector embeddings, and database records processed through enterprise zero-data-retention (ZDR) endpoints are never utilized to train, fine-tune, or reinforce public foundation models (including OpenAI, Anthropic, or Google Cloud models).
05. Anonymized Benchmarking & Research Rights
ProjectAI reserves the right to compile and utilize aggregated, de-identified, and anonymized operational telemetry (e.g., generalized workflow latency reductions, average hours saved across industries) strictly for producing industry research benchmarks, ROI indices, and improving proprietary advisory methodologies, without disclosing any client-identifiable or confidential records.
06. Sub-Processors & Third-Party Infrastructure
We partner with industry-standard, SOC2/ISO 27001 certified cloud and communication infrastructure providers (such as AWS, Cloudflare, Supabase, and WhatsApp Business API) to operate our advisory communications. Third parties are restricted from utilizing personal data for any purpose other than providing necessary infrastructure support.
07. Anti-Scraping & Intellectual Property Notice
Automated crawling, scraping, data mining, or extraction of ProjectAI’s interactive simulators, ROI calculators, architectural blueprints, or website copy for any commercial purpose or AI model training without prior written authorization is strictly prohibited.
08. Data Subject Rights & DPO Contact
Subject to verification and statutory exemptions under the PDPA, you may request access to, correction of, or withdrawal of consent for your personal data by contacting our designated Data Protection Officer:
Designation: Data Protection Officer (DPO)
Entity: ProjectAI LLP (Singapore)
Response Window: Within 30 calendar days upon identity verification
gavel
Enterprise Service Terms & SOW Governance
Legally protective engagement terms, AI IP segregation, API dependency disclaimers, and 3-month liability caps.
01. Engagement Scope & SOW Precedence
ProjectAI LLP delivers enterprise AI workflow architecture, private knowledge base deployments, and corporate project management upskilling. Deliverables, payment milestones, and technical specifications are defined in mutually executed Statements of Work (SOW) or Master Services Agreements (MSA). In the event of any conflict between this website's general terms and an executed SOW, the specific SOW shall strictly prevail.
02. Intellectual Property Segregation
- Client Data & Bespoke Deliverables: All client data, raw database records, proprietary vector embeddings, and custom workflow automations built specifically for the Client shall become the property of the Client upon full settlement of contract fees.
- ProjectAI Core Engine IP: ProjectAI retains sole and exclusive ownership of all pre-existing software libraries, architectural frameworks, prompt engineering templates, ROI calculators, and generalized workflow modules, which ProjectAI may freely reuse in other client engagements.
03. AI Probabilistic Output & Hallucination Disclaimer
Large Language Models (LLMs) and Generative AI systems operate probabilistically. While ProjectAI implements strict deterministic validation schemas, prompt guardrails, and automated checks:
- AI-generated classifications, summaries, and responses are provided to augment human productivity and require human-in-the-loop oversight for critical financial, legal, medical, or statutory decisions.
- Calculators, time-saving estimates, and simulation blueprints on this website are illustrative models and do not constitute a contractual warranty of specific commercial outcomes.
04. Upstream API & Model Dependencies
Custom AI pipelines rely on upstream foundation model providers and cloud SaaS platforms (e.g., Anthropic, OpenAI, Google Cloud, n8n, Make). ProjectAI shall not be held liable for third-party API service outages, latency spikes, breaking schema modifications, rate-limiting policies, or upstream model deprecations beyond ProjectAI's reasonable control.
05. Prohibited Uses & Anti-Scraping Protection
You agree not to: (a) reverse-engineer, decompile, or extract source code or simulation algorithms from this website; (b) deploy automated bots, spiders, or scrapers to extract website copy, calculators, or architectural diagrams; or (c) publish public benchmark tests or comparative performance reviews of ProjectAI's solutions without prior written authorization.
06. Delivery, Hypercare & Maintenance Boundaries
Following user acceptance testing (UAT) and deployment, ProjectAI provides a complimentary hypercare stabilization window as specified in the SOW. Requests for new integrations, workflow modifications, or adjustments necessitated by third-party tool changes after hypercare require a separate Change Order or an ongoing Care & Maintenance Agreement.
07. Limitation of Liability
To the maximum extent permitted by Singapore law, ProjectAI LLP’s aggregate financial liability arising under or in connection with any engagement (whether in contract, tort, negligence, or otherwise) shall be strictly capped at the total professional fees actually received by ProjectAI from the Client under the applicable SOW in the three (3) months preceding the event giving rise to liability. In no event shall ProjectAI be liable for loss of profits, business interruption, or consequential damages.
08. Governing Law & Dispute Resolution
These Terms, Statements of Work, and all related matters shall be governed exclusively by and construed in accordance with the laws of the Republic of Singapore. Any dispute arising out of or in connection with these Terms shall be subject to the exclusive jurisdiction of the Courts of Singapore.
lock
Enterprise Security Architecture
Bank-grade encryption, tenant isolation, and strict access controls.
vpn_key
TLS 1.3 & AES-256
All API data in transit is encrypted using modern TLS 1.3 cipher suites, with AES-256 encryption at rest.
cloud_done
Tenant Isolation
Private vector stores and RAG embeddings are isolated within client VPCs or dedicated multi-tenant schemas.
security
Zero Retention Endpoints
Foundation model calls route exclusively through commercial zero-logging enterprise tiers.
history
Audit & Access Controls
Role-based access control (RBAC) and immutable logging for administrative traceability.
Compliance Inquiries
For custom security questionnaires or SOC2 / ISO compliance documentation, contact our security advisory team at info@ProjectAI.Asia.