As at: 1 May 2026

Introduction

University-industry collaborations routinely generate complex confidentiality challenges, especially when student dissertations, theses, and academic publications are involved. This updated practice note provides a practical, South Africa-aligned framework for distinguishing between disclosed and derived confidential information, managing publication risk, and ensuring that academic freedom and industry trust remain in balance.

1. Disclosed Confidential Information

Disclosed confidential information refers to information explicitly provided by the industry partner under confidentiality terms.

Characteristics

• Provided directly by the partner in documents, presentations, meetings, or data transfers.
• Clearly marked or identified as confidential (e.g., Confidential, Proprietary).
• Protected under NDAs, collaboration agreements, or institutional confidentiality clauses.
• May include personal information regulated by POPIA when datasets contain identifiable individuals.
• Recognised under the IPR Act and NIPMO Guidelines as part of the partner’s background IP or proprietary know-how.

Examples

• Technical specifications of proprietary equipment
• Business strategies, pricing models, client lists
• Internal R&D data or experimental results generated by the partner

Rule of Thumb

If the partner hands it over, labels it, or communicates it under confidentiality, it is disclosed confidential information.

2. Derived Confidential Information

Derived confidential information is created by students or researchers, but still reveals or depends on the partner’s confidential inputs.

Characteristics

• Not directly provided by the partner, but inferred, calculated, modelled, or deduced from disclosed information.
• Arises from experiments, simulations, or analyses using partner data.
• Can expose trade secrets even when not explicitly marked as confidential.
• Often the most contentious category in publication review processes.

Examples

• Performance results of catalysts tested with proprietary formulations
• Process models built using confidential operational parameters
• Comparative analyses that reveal trade secrets by implication

Rule of Thumb

If publication would allow a third party to reconstruct, estimate, or reverse-engineer the partner’s confidential information, it is derived confidential information.

3. Publication Review & Confidentiality Management Process

3.1 Contractual Foundations

Collaboration agreements should:

• Define disclosed and derived information with examples.
• Specify review rights and timelines.
• Clarify consequences of non-response.
• Align with POPIA, the IPR Act, and NIPMO’s 2024-2026 guidance.

3.2 Two-Stage Publication Review Process

Stage 1 – Technical Accuracy Review (Partner)

Timeline: 10 working days
Partner checks:

• Technical correctness
• Misinterpretations
• Safety-sensitive content
• Reputational risk

Stage 2 – Confidentiality Review (Partner + University)

Timeline: 10 working days
Reviewers identify:

• Disclosed confidential information
• Derived confidential information
• POPIA-regulated personal data
• Trade-secret-sensitive results

If the Partner Does Not Respond

Include a clause such as:

If the partner does not respond within the agreed review period, the university may proceed with publication, provided reasonable steps were taken to identify and remove confidential information.

Dispute-Resolution Mechanism

If the partner and university disagree on redactions:

• Escalate to the research office.
• Refer to the contract’s dispute clause.
• Use a neutral technical expert if required.

4. Segmentation, Anonymisation & Aggregation Techniques

4.1 Segmentation

Separate:

• General academic insights (publishable)
• Partner-specific details (restricted)

4.2 Practical Anonymisation Techniques

Examples students can use:

• Replace exact values with ranges (e.g., “between 120-150°C”).
• Normalise data (e.g., “relative efficiency index”).
• Mask variables (e.g., “Catalyst A”, “Process Line X”).
• Aggregate results (e.g., “average performance across three trials”).
• Remove or generalise proprietary parameters.

5. Institutional Responsibilities

5.1 Supervisors

• Train students to recognise derived information.
• Ensure early segmentation of confidential vs. publishable content.
• Guide students on anonymisation and aggregation techniques.
• Ensure compliance with POPIA when handling personal data.

5.2 Research Office / Technology Transfer Office (TTO)

• Provide templates for confidentiality segmentation.
• Maintain standard NDAs and collaboration agreement clauses.
• Support publication review workflows.
• Advise on IP ownership, student IP, and partner IP boundaries.

6. Student IP vs. Partner IP

6.1 Student-Generated IP

Students typically own the IP they create unless:

• They sign an IP assignment agreement.
• The university’s IP policy assigns ownership to the institution.
• The work is funded under a contract that assigns IP to the partner.

6.2 Partner IP

Partner background IP and confidential information remain the partner’s property.

6.3 Interaction with Derived Information

Derived information may:

• Be student-generated IP, but
• Still be confidential to the partner
• And still be subject to publication restrictions

This distinction must be made explicit in contracts and supervision.

7. Diagnostic Checklist for Students

8. Use-Case Scenarios

Scenario 1: Engineering Thesis

A student models a heat-exchange system using proprietary operational parameters.

• The model is student-generated IP.
• The parameters are disclosed confidential information.
• The model outputs are derived confidential information.

Scenario 2: Business Case Study

A student analyses a partner’s pricing strategy.

• Raw pricing data = disclosed confidential information.
• Sensitivity analysis = derived confidential information.
• General pricing theory = publishable.

Scenario 3: Chemistry Experiment

A student tests catalysts provided by a partner.

• Catalyst composition = disclosed confidential information.
• Performance curves = derived confidential information.
• Methodology = publishable.

Conclusion

Understanding and managing the distinction between disclosed and derived confidential information is essential for successful university-industry collaborations. Clear contractual terms, structured review processes, and institutional support ensure that academic research can thrive without compromising industry confidentiality.

Disclaimer: This practice note provides general guidance and should not be considered legal advice. For specific contractual or legal concerns, consult a qualified professional.