Lipstip: the French startup that protects European brands
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
A Birth to Bridge the European Technological Gap
In 2023, the city of Pau witnessed the birth of Lipstip, a startup founded by Nicolas Girardin, Alexander Lerbs, and Nicolas Vonthron. This initiative arose from an alarming awareness regarding Europe's technological competitiveness. Nicolas Vonthron, in an interview with Presse-citron, emphasizes that Europe and the United States shared a similar GDP twenty years ago. Today, the gap has widened, primarily due to technology. According to him, Europe has missed the technological boat for two decades, resulting in a critical delay in the protection of innovations.
In Europe, the intellectual property market faces a paradox: while 90% of SMEs do not protect their intangible assets, those that do see their revenue per employee increase by an average of 44%. Nicolas Vonthron explains that this is not due to a lack of willingness, but rather a logistical capacity issue. With a 10% increase in trademark filings last year, specialized firms are overwhelmed, suffering from a shortage of professionals and numerous retirements.
An Innovative Approach with Small Language Models
Lipstip stands out by refusing to follow the trend of large language models, such as those developed by OpenAI or Anthropic. The startup does not engage in generative AI, as these models, while impressive for certain tasks, rely on statistical approximations that can lead to legal errors. To meet the demands of lawyers and industrial property advisors, Lipstip has designed its own ultra-specialized mathematical models, called Small Language Models (SLM).
These models were developed in collaboration with the University of Pau and the Pays de l'Adour (UPPA) and CY Tech. Lipstip's platform automates the comparison of products and services by analyzing phonetic, visual, semantic, and conceptual criteria. It relies on a vast database of case law comprising hundreds of thousands of decisions from the National Institute of Industrial Property (INPI) and courts.
Nicolas Vonthron illustrates that these models are specialized for a single task, but they execute it perfectly. While generalist AIs achieve an accuracy of 40 to 50%, Lipstip claims an accuracy exceeding 95%. Moreover, the startup systematically cites its sources, and thanks to its lightweight architecture, it benefits from energy frugality. The models can operate locally on a simple desktop computer, thus avoiding the energy-intensive infrastructures of data centers.
Significant Time Savings for Professionals
Lipstip's tool transforms the work of intellectual property professionals, particularly during opposition procedures. These processes, aimed at preventing the filing of overly similar trademarks, were previously tedious and time-consuming. Thanks to the platform, experts can now save an average of three hours per case.
Beta testers, who have been using the solution for a year, report productivity gains of up to 70%. By freeing up "brain time" for experts, Lipstip hopes to create a positive domino effect for the economy. By reducing processing costs, firms can make trademark law more accessible to small and medium-sized enterprises in Europe.
A SaaS Subscription-Based Business Model
Lipstip adopts a classic deeptech business model, based on an annual SaaS subscription per user. The startup's main argument is unlimited access, made possible by the lightweight nature of its proprietary models. Unlike AI giants that charge based on usage, Lipstip offers complete budget visibility to its clients.
A Digital Sovereignty Challenge
Beyond technological innovation, Lipstip is waging a political battle for digital sovereignty. The platform, fully developed and hosted in Europe, ensures that the strategic data of European innovations remains under European jurisdiction. Nicolas Vonthron warns against the current systemic risk in industrial property, where most technical tools are under extra-European control or incorporate American AIs, creating a dangerous dependency.
Marketed for only a month, the startup from Pau has clear ambitions for the future. After consolidating its database in France and Europe, it plans to expand into the Benelux and Germany. Lipstip also intends to integrate automated monitoring features to cover the entire lifecycle of trademarks. The goal is to become the new standard of sovereignty for protecting ideas in Europe.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.