Faster to breed, harder to own: what the EU's NGT deal means for plant IP

The headlines on the European Union's new genomic techniques deal are about science, sustainability and food security. For anyone who breeds, owns or commercialises plant varieties, the decision that will shape the next decade sits quietly in the fine print, in how the deal handles patents and plant variety rights.

What the deal actually changes

In December 2025, the Council and the European Parliament reached a provisional agreement on plants developed with new genomic techniques, the gene-editing methods such as CRISPR that change a plant's own DNA. Subject to ratification expected during 2026, the rules would start to apply from 2028.

The agreement splits these plants in two. NGT1 plants, those carrying small edits that could also have arisen naturally or through conventional breeding, would be treated like conventional plants and freed from the strict GMO regime. NGT2 plants, with more complex changes, stay under existing GMO rules, including labelling. Organic production still excludes both. The European Commission frames the change around resilience and food security, and the breeding case is real: Wageningen University & Research has put the time saving from CRISPR at eight to ten years against conventional crossing for some traits.

That is a genuine acceleration. It is also where the commercial question starts.

The part that matters for IP is in the fine print

Read past the science and you find the part that will occupy breeders and rights holders: patents.

Under the provisional deal, edits made with new genomic techniques stay patentable. To register an NGT1 plant, a company must disclose its existing and pending patents in a public database. Patent holders can voluntarily flag a willingness to license. An expert group, drawing in the European Patent Office and the Community Plant Variety Office, will examine how patents affect NGT plants. And a year after the rules take effect, the Commission will study what patenting does to innovation, to the availability of seed for farmers, and to the competitiveness of European breeding.

Every one of those measures exists for one reason. The EU knows the real contest over gene-edited varieties will be fought over intellectual property, not safety. This was hard-won. The Parliament had pushed to restrict or even ban patents on gene-edited plants, and the final framework keeps them patentable while trying to manage the risk through transparency instead.

Why patents and plant variety rights pull in different directions

This is where breeders get caught, because the two systems that can protect a new variety behave very differently.

Plant variety rights protect a specific variety, the whole combination of characteristics, and they carry a breeders' exemption. Anyone can take a protected variety, breed something new from it, and commercialise the result without asking the original holder, essentially derived varieties aside. That open exemption is the engine that keeps breeding moving.

A patent works the other way. It protects a trait or a technique rather than a finished variety. In Europe, plants produced by essentially biological processes, ordinary crossing and selection, are increasingly shut out of patentability, but a trait introduced by CRISPR counts as technical, so it stays patentable. The practical effect is sharp: you can study a patented trait under a narrow research exemption, but you cannot sell a new variety that contains it without a licence. The full breeders' exemption shrinks to a limited one. You may breed; the product can still be blocked.

For a breeder raised on the open culture of plant variety rights, that is a different game. A patent is published for all to see, so there is no quiet head start, and as trait patents multiply they form thickets, overlapping rights that can close off a breeding path or drive up the cost of getting clear.

What to do before 2028, not after

The rules land in 2028. The positioning happens now. Five things are worth doing early.

• Run freedom to operate before you commit. Check whether a trait, or the method behind it, is patented before you build a programme on it. A variety that depends on a patented trait you have not cleared is a liability wearing the costume of an asset.
• Choose your protection mix on purpose. Plant variety rights, a patent, trade secrets around the data and method, or some combination. The new disclosure database changes the maths, because your filings, and your competitors', become visible.
• Plan for split ownership. The trait and the variety may belong to different parties. Build licences and commercial terms for that from the start, including access and royalties, rather than discovering it at launch.
• Know the access routes. Mechanisms such as the International Licensing Platform for vegetables exist to give members access to patented traits on fair terms through arbitration, and national steps like the Netherlands' limited breeders' exemption in patent law point the same way. Work out which apply to you.
• Keep the rest of the stack in view. Access and benefit-sharing under the Nagoya framework still applies to the genetic resources you use. The patent question does not travel alone.

The science gets the headlines, the IP decides the winners

New genomic techniques make varieties faster and cheaper to create. They also make the question of who owns what a great deal sharper. The breeders and rights holders who come out ahead will be the ones who treated that as a strategy question early, while the rest of the room was still reading the safety debate.

At Greenstone we work with breeders, IP holders and the companies that take varieties to market on exactly this: freedom to operate, the balance between patents and plant variety rights, and licensing that holds up in a real market. If new genomic techniques are on your roadmap, the IP strategy is worth starting before the regulation does.