Insights

The Japan Patent Office has now moved from broad design-policy signalling to something much more useful for applicants: supplementary explanation tied to its Design Examination Guidelines and a first set of practical case examples focused on partial designs and graphical user interfaces under Hague filings. For teams designating Japan, the message is fairly direct. How the claimed portion is isolated, how unclaimed matter is shown, how a GUI sequence is organized, and how much explanatory detail is built into the filing will all matter more than many applicants would like to admit.

What makes this development important is not that Japan has suddenly changed course on design protection. It is that the JPO is tightening the operational bridge between existing examination rules, Hague-based international design practice and the day-to-day realities of review. That changes the strategic question. The issue is no longer simply whether a portfolio should use the Hague route for Japan, but when a single global representation set is still good enough and when Japan needs a more deliberately localized filing package.

IPOS has used its supplemental AI patent guidance to make one point much harder to ignore: for AI and data-heavy inventions, the fight is often no longer about whether software-related subject matter can be patented in the abstract. It is about whether the claim and the specification show a real technical contribution tied to a specific problem, rather than a mathematical model dressed up with industry language.

That is why the latest clarification matters. The practical pressure is now falling on the parts of an application that many teams still treat too loosely: training-data processing, dataset improvement, post-training model adjustment, and the disclosure needed to show why those steps produce a technical result. If the claim reads like a generic algorithm, or like ordinary hardware merely running a model, the old “mathematical method” risk has not gone away. IPOS is effectively asking applicants to make the chain visible: what the model does, what inputs and outputs matter, what technical setting is being addressed, and what result follows from that design.

IP Australia has signalled a more deliberate effort to make trade mark opposition practice less expensive and less drawn out, with one of the most notable developments being a trial of Early Neutral Evaluation (ENE) before parties are pulled deep into the full evidence cycle. On the model now being discussed, a trade mark applicant and an opponent may choose to place their preliminary materials before an IP Australia decision-maker and receive a non-binding view on the relative strength of the case, the real points in dispute, and whether settlement or withdrawal makes more commercial sense than a long procedural fight. The idea is not to replace the opposition system, but to stop weaker or more negotiable disputes from automatically sliding into the costliest phase.

That matters because in trade mark oppositions, cost and momentum often drive behaviour as much as legal merit. Once evidence rounds begin, even parties with a shaky position may keep going simply because too much time and money has already been spent. If ENE works as intended, it could change that psychology early. Applicants would need to prepare their distinctiveness story, commercial use context and filing rationale sooner, while opponents would face more pressure to show real substance rather than rely on procedural drag. For businesses watching Australia, the practical message is straightforward: opposition strategy may start shifting from endurance to early case assessment, and that usually rewards parties who know their record before the fight hardens.

China’s CNIPA has issued a new work briefing on deeper action against bad-faith trade mark filings and paired it with a group of guiding cases on “rapid invalidation.” The practical signal is stronger than a routine policy reminder. For large-scale stockpiling, free-riding on established brands and opportunistic hot-topic filings, the system is not only promising stricter control in principle. It is accelerating case handling and tightening the evidentiary threshold in ways that can change how disputes are won or lost.

The more consequential part may be what the briefing says about three-year non-use cancellation practice. For a long time, some registrants could still survive by piecing together invoices, contracts, sample photos and scattered paperwork that suggested some form of use without really proving sustained market activity. CNIPA is now indicating a much narrower tolerance for that approach. Merely listing a mark name on an invoice, producing token shipment records or staging low-intensity use designed mainly to preserve registration is becoming a weaker and weaker answer if the evidence cannot show genuine commercial circulation.

With the April 2026 EPO Guidelines now in force, practitioners this week have been focusing on three changes with immediate drafting and prosecution relevance. First, the EPO has formally integrated colour and greyscale drawings into practice for electronic filings, moving beyond the old black-and-white-only baseline. Second, the Guidelines make clear that applicants and representatives remain fully responsible for everything filed with the EPO, even where patent applications or replies have been prepared with the assistance of artificial intelligence. Third, the revised text now reflects Enlarged Board decisions G 1/24 and G 1/23, sharpening the treatment of claim interpretation and the prior-art status of products already placed on the market.

These are not isolated edits. Together they point to a more evidence-driven and accountability-heavy practice environment. Colour drawings will help in areas such as interfaces, materials and complex structures, but they also make consistency, contrast and amendment discipline more important. AI can still support drafting, but it no longer leaves room for anyone to treat machine-generated text as a compliance buffer. And with G 1/24 and G 1/23 now expressly embedded in the Guidelines, applicants would be wise to review how they define claim terms, document product features and position prior art from the outset rather than trying to repair those issues later in examination or post-grant disputes.

Since 1 April 2026, the UK Intellectual Property Office has raised official fees across patents, trade marks and designs, with the IPO describing the overall increase as averaging about 25%. The online fee for a UK trade mark application has moved from £170 to £205, while the patent search fee has risen from £150 to £200. This week the IPO also continued its transition messaging through fresh One IPO roadmap updates. The more consequential trade mark development, however, is that the government has confirmed the removal of the series marks service, while allowing existing registrations to remain on the register.

That change matters more than the loss of a peculiarly British filing convenience. Consultation materials and the government response point to repeated misunderstanding of the series mechanism, high objection rates for unrepresented filers, and concern that some applicants were using it to obtain multiple examinations at reduced cost. UK trade mark practice is therefore moving closer to the international norm of separate applications for distinct variants. Brand owners that used to bundle minor logo, layout or word-mark variations into one filing now need a sharper plan on which versions truly merit standalone protection.