Korea now punishes trade-secret brokers, too — what the amended Unfair Competition Prevention Act means for your lab notebook

Korea's National Assembly passed an amendment to the Unfair Competition Prevention and Trade Secret Protection Act on August 20, 2026. Two changes stand out: brokering, introducing, or inducing trade-secret theft is now its own category of infringement, and hacking is now explicitly named as a method of unlawful acquisition. Until now, the "broker" — the person who arranges the leak without ever touching the information directly — was often the hardest actor to reach under the old law. This post summarizes what changed and why it connects directly to how you keep research records.

What's changing

The old law was built around people who directly acquired, used, or disclosed a trade secret. The problem is that the pattern investigators actually see doesn't always fit that mold — a broker who approaches key staff and arranges for them to "bring materials along" when they change jobs, often posing as a recruiter, never personally acquires or uses the information. That left a real gap in enforcement.

The amendment closes it as follows.

Coverage of the amendment describes the same recurring pattern behind it: key employees are repeatedly recruited away with a broker arranging the move, yet the law had no clause that reached the broker directly. That said, the exact penalty ranges (prison terms, fine ceilings) and the effective date are still being finalized through the promulgation process. Don't rely on this summary alone — check the official statutory text and the Ministry of Government Legislation's notice on the amendment for the promulgation date, effective date, and exact wording.

Punishing the broker doesn't excuse you from proving it was your trade secret

This is where the amendment connects to research records. Being able to prosecute a broker doesn't lighten the burden of proof on the company that was harmed. Whether the case is criminal or civil, it starts the same way: you first have to prove when, and in what form, the information became your trade secret before anyone can be held liable for brokering, inducing, or unlawfully acquiring it.

To qualify as a trade secret under the Act, information has to meet three tests — it must not be publicly known, it must have independent economic value, and it must have been kept reasonably confidential. R&D records are usually the most common evidence for the latter two. When a dispute turns on what a departing employee actually took, and whether that information was already yours before they took it, your lab notebook has to prove two things at once:

• that the technology or data existed inside your organization **before a

specific point in time**, and

• that it hasn't been altered since — in a form a third party can

verify, not something inserted or backdated after the dispute began.

If either of those is shaky, it doesn't matter how solid the new provisions against the broker are — the case falls apart at the first step, before anyone gets to brokering or inducement at all.

Where these cases actually get won or lost

The issues that come up again and again in trade-secret leak disputes are fairly consistent.

• The gap between the resignation date and the record date — if a

project's records go quiet for months before someone resigns, or get suspiciously tidied up right before they leave, it becomes hard to show a court what was genuinely your trade secret in the first place.

• The limits of an internal server timestamp — creation and

modification times stamped by an in-house system are never fully free of the suspicion that an administrator could have changed them. Once the other side argues the date was set arbitrarily by the company, your own logs alone won't rebut that.

• The gap between the lab notebook and the actual working files — if

the notebook only holds a summary while the real source code or design data lives in a separate repository, the notebook can't prove when the trade secret actually existed. The underlying material itself needs its own timestamp.

These are the same issues we covered in what counts as a valid research note and in proving prior use and trade-secret ownership. Now that brokers and intermediaries are within reach of the law too, the fight is likely to shift even further toward one question: not "can we punish who took it," but "can we prove we had it first."

Self-check

• [ ] Is the moment your core technology or know-how gets recorded

verifiable by a third party — or does it rely only on an internal server log?

• [ ] In roles with frequent turnover, are there gaps in the record —

entries written in a batch, or backdated after the fact?

• [ ] Is material you'd claim as a trade secret only summarized in your

notebook, with the actual originals sitting in a separate system with no timestamp of their own?

• [ ] Do you have access-control and security-policy documentation to back

up "reasonable confidentiality," kept separately from your R&D records?

This post summarizes news coverage. The exact statutory language, effective date, and penalty levels should be confirmed against the promulgated text once it's finalized. If you're dealing with an active leak dispute or response strategy, consult a qualified attorney.