Just outside Lausanne, Switzerland, Ioannis Ierides, a business-development manager at Iprova, faced the challenge of showcasing his company's AI-driven idea generation in a brief presentation. Iprova specializes in providing patentable invention proposals that leverage AI to analyze literature across various fields and suggest innovative combinations of existing concepts. Their clients include major corporations like Procter & Gamble and Panasonic. During our meeting, Ierides illustrated how Iprova's software identifies potential inventions. After inputting relevant keywords, the tool used semantic-search capabilities to produce a pie chart that visually represented research areas. Ierides then refined the search, generating clear new invention proposals, such as a combination of advanced microphones and software for speaker identification. While this example appeared straightforward, Iprova’s actual process involves deeper iterations and cross-field connections to create more complex inventions. Iprova co-founder Julian Nolan emphasizes the critical nature of invention, asserting that being first to market is essential. The company often identifies cutting-edge ideas, exemplified when Panasonic sought novel applications for autonomous vehicles.

The software proposed turning these vehicles into gig workers by providing delivery services, leading to a patent application. AI technology is not new in patent discovery; however, the integration of advanced generative AI models enhances Iprova's capacity to summarize vast quantities of technical texts, enabling them to track emerging innovations effectively through a constantly updated resource called "the Index. " Iprova's inventions often intersect different fields, evident in a project connecting lithium batteries and message encryption for Ericsson. The software linked the random degradation of lithium batteries to unique encryption key generation, resulting in multiple patents. Yet, not all patents result in actionable inventions, as companies may use them to safeguard intellectual property rather than bring products to market. Iprova relies on client collaboration to generate market-ready ideas, avoiding a model based solely on acquiring patents for licensing. The increasing complexity of regulations in areas such as telecommunications has prompted Iprova to develop tools that help decipher industry standards, aiding inventors in creating compliant inventions. Despite the advancements, both Iprova and industry experts acknowledge that human insight remains vital in assessing novelty and applicability in inventions. As AI evolves, the partnership between human creativity and machine capabilities will be critical in shaping the future of innovation. AI's role may eventually shift from mere assistance to a more integral part of the invention process. Ultimately, the interplay of AI and human ingenuity is not just about creating inventions but understanding their origins and potential. The synergy of these elements may lead to predicting competitors' innovations, fostering a new era of inventive collaboration.