Partnership integrates Patsnap’s proprietary global patent dataset with Intellegens’ Alchemite Suite, giving R&D teams access to patentability and freedom-to-operate checks during experimental design.
Cambridge & London, UK – 25 June, 2026 – Intellegens, a pioneer in advanced machine learning for R&D, and Patsnap, the global leader in AI-native innovation intelligence for IP and R&D, today announced a strategic partnership connecting Patsnap’s global patent dataset with Intellegens’ Alchemite™ Suite. The partnership is designed to give R&D teams in chemicals, materials, and FMCG access to patent insight earlier in the development process, at the point of experimental design rather than after testing is complete.
Formulation development and IP review have traditionally operated as separate functions. Scientists optimize a candidate material or chemical formulation and then submit it for patent clearance, a sequence that can result in weeks of experimental work being forfeited or redirected when a conflict is identified.
Alchemite uses machine learning to predict optimal formulations from sparse experimental data, compressing development cycles by 50 to 80 percent by directing R&D teams toward the most promising candidates. That compression matters: according to Patsnap’s 2026 R&D Benchmark Report, based on responses from 200+ R&D professionals, more than a third of organizations (37%) spend between 25% and 40% of their entire R&D budget on projects that never reach market, and when projects are terminated late in development or testing, 45% of respondents estimate the wasted investment at $1 million or more per project.
Through the partnership, Intellegens’ AI agent, Alchemite Insight, can now validate candidates against Patsnap’s database of more than 170 million global patent records for novelty and freedom-to-operate before bench work begins. Where a conflict exists, the system can surface alternative ingredients or process parameters that maintain the performance target while staying clear of protected intellectual property.
The integration runs through Patsnap’s Open Platform, which allows AI systems and agents to call its IP intelligence programmatically. This is what allows IP validation to sit natively inside Alchemite’s design loop rather than as a separate, manual step.
“Patsnap’s R&D customers have consistently flagged the gap between formulation work and IP review as a drag on their go-to-market timelines. This partnership puts IP intelligence where it can influence decisions at the design stage, not after the fact,” said Ray Chohan, Co-Founder at Patsnap.
“By embedding Patsnap’s IP intelligence directly into the Alchemite experimental design process, we remove the friction between invention and legal validation,” said Ben Pellegrini, CEO of Intellegens. “Scientists don’t just get a high-performing formula; they get the confidence that it is a viable, legally clear asset from day one.”
About Patsnap
Founded in Singapore in 2007, Patsnap is a global leader in innovation intelligence. Leveraging domain-specific AI agents and advanced analytics, Patsnap empowers IP and R&D teams to uncover insights, accelerate discoveries, and protect breakthroughs. Trusted by more than 15,000 enterprises, law firms, and research institutions across over 50 countries — including one-third of the world’s largest R&D spenders — Patsnap transforms how organizations innovate and make critical decisions. Learn more at www.patsnap.com.
Patsnap media contact: Kate White, media@patsnap.com.
About Intellegens
Intellegens provides unique machine learning software, Alchemite™, designed to solve complex problems with sparse or noisy data. It accelerates the development of novel materials, chemicals, and processes across industries ranging from aerospace to pharmaceuticals. For more information, visit intellegens.com.
Intellegens media contact: Caroline Osborne, info@intellegens.com.
Supporting image
Patsnap in Alchemite – the Alchemite Insight AI is being used to explore information about potential formulations proposed for experiment by the Alchemite machine learning method. It returns information on patent status of these candidates, informed by the Patsnap database. Click for larger image.
