AI-deation

Identifying and developing innovative research ideas is time-consuming, resource-intensive, and often limited by human capacity to synthesize vast, scattered information. AI-deation focuses on two main use cases: recombining existing knowledge into new applications and creating completely novel research ideas. The project aims to deliver a practical, privacy-conscious, and cost-effective prototype, ultimately contributing towards semi-automated ideation support for industrial research and innovation.

Project description

In today’s fast-paced innovation landscape, identifying high-impact research opportunities and use cases is increasingly difficult. Traditional ideation processes are time-consuming, dependent on expert availability, and limited in their ability to harness and recombine the vast amounts of available technical and interdisciplinary knowledge. Enterprises need scalable, cost-effective tools to accelerate innovation without compromising data privacy or sustainability.

AI-deation is a joint initiative by Siemens and fortiss to develop an intelligent ideation platform powered by Large Language Models (LLMs) and privacy-friendly Small Language Models (SLMs). The system enhances human creativity by generating, refining, and validating research ideas and technical use cases. It integrates seamlessly with Siemens’ existing processes to support knowledge recombination and novel ideation, helping experts move from raw capabilities to impactful innovations—faster and more efficiently.

Research contribution

A working prototype that generates use cases and research ideas based on Siemens’ existing capabilities, supporting two use cases:

Recombine existing knowledge
Create novel ideas beyond what’s currently published or known

Projekt duration

01.02.2025 - 31.12.2025

Your contact

Dr. Severin Kacianka

+49 89 3603522 286
kacianka@fortiss.org

Name	Purpose	Lifetime	Type	Provider
_pk_id	Used to store a few details about the user such as the unique visitor ID.	13 months	HTML	Matomo
_pk_ref	Used to store the attribution information, the referrer initially used to visit the website.	6 months	HTML	Matomo
_pk_ses	Short lived cookie used to temporarily store data for the visit.	30 minutes	HTML	Matomo
_pk_cvar	Short lived cookie used to temporarily store data for the visit.	30 minutes	HTML	Matomo
_pk_hsr	Short lived cookie used to temporarily store data for the visit.	30 minutes	HTML	Matomo