Trade Secrets in Technology Financing

A venture investor is financing a Nanjing laser-technology company whose valuation rests on proprietary manufacturing know-how. The company owns some patents, but core process parameters were developed with a research institute and several founders before incorporation. Management calls all of this 'our IP.' The investor needs to know which rights are actually owned, licensed or protected as trade secrets and which defects must be fixed before closing rather than left to warranties.

Article 10 of the 2025 revised Anti-Unfair Competition Law defines trade secrets through non-public status, commercial value and corresponding confidentiality measures and prohibits specified forms of misappropriation. Article 39 contains important evidentiary burden-shifting mechanisms where the rights holder first provides the required preliminary evidence. Patent law and corporate ownership rules also matter where technology is registered, assigned, licensed or contributed.

The specific issue#

1. Build a core-technology map before valuing the company#

Map the products, process steps, datasets and code repositories that actually drive valuation before debating price. Investors who price “technology company” without a core-technology map later discover that the valuable know-how sits with a founder, institute or unnamed contractor. The map should identify what must be owned or exclusively controlled at closing.

2. Define specific trade-secret points rather than accepting 'know-how' as a category#

Replace vague “know-how” labels with a schedule of specific secret points: formulas, process windows, source modules, customer-adaptation methods or datasets. Chinese trade-secret enforcement rewards specificity. A claim that “all technology is confidential” is usually neither operable for diligence nor credible in litigation.

3. Test whether corresponding confidentiality measures actually exist#

Test corresponding confidentiality measures against each claimed secret: access lists, repository permissions, NDAs, exit certifications and physical or logical segregation. Anti-Unfair Competition Law protection expects reasonable measures. If the company cannot show measures matched to the secret, ownership talk will not survive diligence or a later injunction request.

4. Review access controls, repository records, supplier restrictions and employee exits#

Review repository logs, badge and system access, supplier confidentiality terms and employee-exit checklists. The question is who could copy the secret and whether the company would know. Weak off-boarding of engineers with admin rights is a classic defect that converts a financing into a leakage event.

5. Separate research-institute background IP from startup-created foreground IP#

Separate research-institute or university background IP from startup-created foreground IP. License scope, field-of-use, assignment consent and publication rights often limit what the company can commercialize or pledge to investors. Treat institute arrangements as title documents, not historical footnotes.

6. Check founder-owned patents and pre-incorporation technology#

Check patents and technology still registered or held in founder or affiliate names. Pre-incorporation inventions that never moved into the company create enforcement and disclosure problems in later rounds. Require assignment agreements with chain-of-title evidence before treating those assets as company-owned in the valuation model.

7. Review employee inventions, assignments and remuneration arrangements#

Review employment invention assignment clauses, remuneration policies and actual payment practice. Paper assignments that conflict with mandatory inventor-remuneration rules, or that were never signed by key engineers, leave gaps exactly where investors assume clean ownership.

8. Investigate whether any former employee or competitor has already accessed the secret#

Investigate whether former employees, failed joint-development partners or competitors already possess the claimed secret. Prior leakage changes both enforcement prospects and the investor’s leverage. If a parallel product already embodies the secret, the financing thesis may need a different IP theory than exclusivity.

9. Use Article 39's evidentiary logic to test whether the company could win an enforcement case#

Stress-test the file as if the company had to seek evidence preservation or prove infringement under Article 39’s evidentiary logic. If the company cannot identify the secret points, measures, ownership and suspected pathway of misappropriation, it does not yet have an investable enforcement asset.

10. Check whether patent publication has already disclosed information claimed as secret#

Compare patent publications and public filings against information still claimed as trade secret. Once technical content is disclosed in a patent document, secrecy claims over the same content collapse. Investors should not pay trade-secret value for matter the company has already published.

11. Review open-source and third-party software separately from proprietary ownership#

Review open-source and third-party components separately from proprietary modules. GPL or other copyleft obligations, missing licenses and mixed repositories can force source disclosure or block commercialization. Ownership of proprietary code does not cure an open-source compliance defect in the same product.

12. Classify every IP defect as pre-closing remediation, covenant, indemnity or valuation issue#

Classify each IP defect as pre-closing remediation, closing condition, post-closing covenant, indemnity, representation or valuation adjustment. Mixing all defects into a single “IP cleanup” covenant hides which rights are indispensable to the product the investor is funding.

13. Require assignment or durable licensing where core technology is held outside the company#

Where core technology remains outside the company, require assignment or a durable exclusive license with recordable terms before closing. A soft undertaking to “transfer later” leaves the investor funding a business that does not control its crown-jewel rights.

14. Use staged data-room disclosure to avoid destroying secrecy during financing#

Use staged data-room access so deep technical disclosure happens only under tightened NDAs, watermarking and need-to-know lists. Financing diligence that dumps full process recipes to every bidder can destroy the secrecy the valuation assumes.

15. Preserve version history and technical evidence supporting ownership#

Preserve version history, commit logs, lab notebooks and inventor declarations that prove when and by whom core technology was created. In a later dispute or IPO diligence, chronology often matters as much as the current repository state.

16. Use specific conditions precedent for indispensable technology rights#

Make indispensable technology rights—assignments, institute consents, key employee acknowledgments—conditions precedent with clear document lists. If those rights cannot be delivered, the investor needs a walk right or a redesigned deal, not a vague best-efforts covenant.

17. Create post-closing IP governance rather than treating diligence as a one-time exercise#

Install post-closing IP governance: access reviews, invention reporting, exit certifications and vendor confidentiality audits. Diligence is a snapshot; without governance, the same leakage patterns reappear before the next financing.

18. Protect investor information rights so they do not cause uncontrolled technical disclosure#

Draft investor information and inspection rights so they do not compel uncontrolled technical disclosure to affiliates, co-investors or external advisors beyond agreed protocols. Information rights should not become an accidental dissemination channel for the secrets backing the investment.

19. Plan for the next financing or IPO because the same defects will reappear#

Assume the next financing or IPO will re-examine the same title, secrecy measures and inventor issues. Temporary workarounds that “get this round closed” often become disclosure or remediation crises under public-market standards.

20. Do not close until the investor can identify the legal right supporting each core product#

Do not close until the investor can point, for each core product feature in the technology map, to the contract, assignment or license that confers the legal right to use and enforce it. If that link cannot be shown, the financing is pricing an asset the company may not own.

Worked scenario: a financing that depends on technology the company may not fully control#

Assume an investor is considering a Series C financing in a Nanjing laser-technology company. Management says the company owns its “core optical process.” Diligence shows a more complicated picture. One patent was filed by the founder before incorporation. A university research institute licensed background technology to the company for five years. The most valuable manufacturing parameters are kept as know-how rather than patented, but access to the process folder is granted to most engineers and several suppliers.

The investor should split the problem into registered IP, contractual rights and trade secrets. The founder-owned patent can be addressed through assignment or a durable license, but the investor should also confirm whether the founder had obligations to a former employer when the invention was made. The research institute agreement should be reviewed for exclusivity, improvement ownership, sublicensing, termination and change-of-control consequences. A five-year license may be insufficient if the company's valuation assumes long-term exclusive commercialization.

The know-how presents a different diligence test. The company should identify specific secret points rather than describe the whole production process as confidential. Counsel should ask where each parameter is stored, who can access it, whether suppliers are contractually restricted, and what happens when an engineer leaves. Under the 2025 Anti-Unfair Competition Law, corresponding confidentiality measures are part of the definition of a trade secret. Weak controls therefore affect both current protection and future litigation strength.

The investor should then classify remedies. If the founder patent is indispensable, assignment can be a condition precedent. If the research license can be extended before closing, an amended agreement may be required. If access controls are weak but fixable, the company can implement a documented trade-secret program before or shortly after closing, backed by a covenant and milestone. A broad IP warranty is not a substitute for any of these steps.

The financing documents should also anticipate the next transaction. If the company expects an IPO or strategic sale, today's investor should require an ownership and protection structure that will survive later diligence. Otherwise the same problem will reappear at a higher valuation and with less time to fix it.

Action checklist before implementation#

1. Confirm who has authority to waive an IP closing condition or accept title risk.
2. Assemble the core evidence set: technology map, secret-point schedule, confidentiality measures, assignments, institute licenses, repository access logs and open-source inventory.
3. Identify the one title defect that would make the product un-investable if uncured.
4. Quantify valuation impact of missing assignments or weak secrecy measures.
5. List third parties whose signatures are required—founders, institutes, key employees.
6. Define the fallback if an indispensable right cannot be delivered by closing.
7. Reconcile data-room disclosure scope with secrecy preservation.
8. Assign owners for each remediation document.
9. Ensure investor rights do not force uncontrolled technical dissemination.
10. Retain a closing file linking each core product feature to a legal right.

Quality-control questions#

Before the matter is closed, an independent reviewer should be able to identify each core product feature and the legal right that supports it. The reviewer should explain why the team avoided treating a management statement that technology is proprietary as a substitute for assignments, secrecy measures and license scope. If that explanation depends on recollection rather than the file, the work is not complete.

Decision tree for the investor#

The investor should begin by ranking technology assets according to business criticality. If the company could still sell its core product after losing a particular patent or license, the defect may be manageable through a warranty or covenant. If loss of the right would stop production or eliminate the company's competitive advantage, the investor should seek actual remediation before closing.

Registered rights should be traced to the current owner and through the chain of title. Founder-owned patents need more than a promise to transfer later if they are central to value. Research-institute licenses should be examined for term, exclusivity, sublicensing, improvement ownership, termination and change of control. A short or revocable license should be reflected in valuation even if no current dispute exists.

Trade secrets require a different analysis because there is no registry. The investor should require the company to identify the secret with reasonable specificity and demonstrate access restrictions, employee confidentiality duties, supplier controls and exit procedures. If every engineer and supplier can access the same process files, the company may have difficulty proving that it used corresponding confidentiality measures.

The investor should also ask whether any patent filing, publication, customer disclosure or conference presentation has already placed the claimed secret in the public domain. Patent and trade-secret strategies should be consistent; a company cannot rely on secrecy for information it deliberately disclosed.

Closing-remediation matrix#

A core founder patent can be assigned as a condition precedent. A research license can be amended to extend term and clarify improvements. Missing employee IP clauses may require confirmatory assignments from current key staff. Weak access controls can be remediated through a documented security and trade-secret program, though the investor should understand that new controls do not retroactively repair every historical weakness.

Known IP disputes should be scheduled separately. The investor should review pleadings, evidence, requested remedies and the commercial consequence of an adverse result. A lawsuit affecting a non-core trademark is different from a trade-secret claim that could stop use of the company's manufacturing process.

The financing agreement should avoid representations broader than the verified facts. If the company relies on licensed technology, the representation should accurately describe that dependence. If an open-source component exists, it should be disclosed rather than hidden under a general statement that the company owns all software.

After closing, the board or management should receive periodic reporting on core IP ownership, trade-secret incidents, license expirations and employee departures involving sensitive technology. That governance reduces the chance that today's diligence defect becomes tomorrow's enforcement crisis.

Evidence that should exist before the investor relies on a trade-secret valuation#

The investor should ask the target to demonstrate how a critical secret moves through the business. For a manufacturing recipe, that means identifying the master file, access permissions, production copies, supplier disclosures, backup systems and employee roles. The company should be able to show that access is narrower than the general employee population and that departures trigger account closure and return or deletion procedures.

The diligence team should also review incidents. If a former engineer joined a competitor, was access preserved, were devices returned, and did the company investigate unusual downloads? A company that has never enforced its controls may still own valuable secrets, but the investor should understand whether the practical protection environment matches the valuation narrative.

Research agreements deserve similar factual testing. If an institute performed foundational development, the investor should examine project proposals, inventor records, patent filings and payment history, not just the final license. Those materials may reveal whether the startup's claimed improvement is actually derived from institute-owned background technology.

Finally, the investment agreement should require continued IP hygiene. The company can covenant to maintain a trade-secret register, restrict access to identified core know-how, obtain confirmatory employee assignments where needed and report material IP disputes to the investor. These are governance controls tied directly to the asset being financed.

Conclusion#

This issue should be managed as a specific legal-control problem. The legal framework must be applied to verified facts and converted into an executable sequence. The central lesson from this scenario is to avoid treating a management statement that technology is proprietary as equivalent to proof of ownership or enforceable trade-secret protection. A strong file shows the legal rule, the commercial decision, the supporting evidence and the fallback if the preferred route fails.

Legal and regulatory sources#

^[1] Anti-Unfair Competition Law of the PRC (2025 Revision): https://www.npc.gov.cn/npc/c2/c30834/202506/t20250627_446247.html ^[2] Patent Law of the PRC — official NPC legal database: https://flk.npc.gov.cn/ ^[3] Company Law of the PRC: https://www.npc.gov.cn/npc/c2/c30834/202312/t20231229_433999.html

This article is general legal information and is not legal advice for a specific matter.