Published Research Can Become Someone Else’s Advantage

23 August 2026

Executive Summary

China’s rapidly advancing robotics industry is producing an unusual lesson about intellectual-property and innovation risk.

A Reuters investigation found that technologies developed through US military-funded university research helped underpin advances later used in China’s highly successful quadruped robotics industry.

Researchers involved in the work told Reuters that designs used by Chinese robotics company Unitree drew heavily on breakthroughs developed through US Army-funded programmes and MIT research, including work behind the Mini Cheetah robot. 

Importantly, Reuters found no wrongdoing in Unitree using the research.

The work had been deliberately published openly to advance scientific knowledge — entirely normal practice in academic research. MIT itself publicly described Mini Cheetah’s design and its intention to share robots with other research laboratories to accelerate experimentation. 

The US Army also openly describes its collaborative research model as partnerships between government, academia and industry designed to advance fundamental robotics research. 

The commercial lesson is therefore more interesting than a conventional intellectual-property dispute:

Information does not have to be stolen to become strategically valuable to somebody else.

UK Impact

UK companies and universities collaborate extensively through:

  • Research partnerships.
  • Government grants.
  • Universities.
  • Joint ventures.
  • Technology accelerators.
  • Supplier development.
  • Industry conferences.
  • Academic publications.

These collaborations are essential to innovation.

But organisations should distinguish carefully between:

knowledge they want the world to learn

and

knowledge that creates their competitive advantage.

Potentially sensitive information can include:

  • Manufacturing techniques.
  • Algorithms.
  • Tolerances.
  • Component architecture.
  • Prototype performance.
  • Materials.
  • Testing results.
  • Process improvements.
  • Supplier identities.

No cyberattack is required if the information has already been legitimately published.

Global Impact

The commercial importance of the technology is becoming increasingly apparent.

Unitree has become one of China’s leading humanoid and quadruped robot manufacturers. Its Shanghai listing this week valued the company at around $50 billion after its shares surged on debut. 

Meanwhile, China’s World Humanoid Robot Games this weekend are showcasing enormous advances in robotic capability and increasingly emphasising real-world industrial tasks rather than simple demonstrations. 

The underlying US Army programme itself was created precisely to stimulate advanced robotics research through cooperation with universities and industry. The Army describes collaborative research alliances as a way of leveraging cutting-edge academic innovation. 

That model brings enormous benefits.

But it also illustrates why organisations need to decide deliberately where open collaboration ends and proprietary knowledge begins.

Our View

Businesses involved in research-intensive sectors should incorporate publication and knowledge-transfer risk into intellectual-property governance.

Companies should ask:

  • Which research can safely be published?
  • Which information provides competitive advantage?
  • Who decides what academics or engineers may disclose?
  • Are patent applications filed before publication?
  • Do collaboration agreements define background and newly created IP?
  • Who owns improvements created jointly?
  • Are conference presentations reviewed before release?
  • Could published technical specifications allow rapid replication?
  • Are suppliers receiving more design information than they require?
  • Do employees understand the difference between confidential know-how and publishable research?
  • Are government-funded projects subject to different disclosure requirements?
  • Is valuable knowledge protected through patents, trade secrets or both?

The objective should not be to stop academic or industrial collaboration.

Open research drives innovation.

The objective is to make publication a deliberate decision rather than an accidental one.

Once technical knowledge becomes public, a company may still own a patent.

But it can no longer make the information unknown.

Sometimes the most important intellectual-property decision therefore happens before anyone presses “publish”.

Risk Indicator: ELEVATED – TECHNOLOGY & R&D

Does This Risk Affect Your Business?

Invictus Risk Solutions helps businesses find practical solutions to insurance, risk and commercial challenges.

From individual businesses to major international organisations, risk is our business.

TALK TO INVICTUS →

Scroll to Top