Hervé
Artificial intelligence, especially conversational AI like the Generative Pre-training Transformer (GPT), is revolutionizing the way organizations collaborate and innovate.
Thanks to an extended version of GPT, trained on an organization's specific data, it is now possible to optimize key processes such as project and partner selection, ideation, and the creation of shared collective knowledge.
How do you do it?
Open Innovation is a concept whereby companies leverage not only their own internal resources and research to innovate, but also ideas, knowledge and technologies from external sources. This can include collaborations with universities, start-ups, industrial partners, customers and other stakeholders.
Un Generative Pre-training Transformer (GPT) étendu dans ce contexte est une version du modèle de langage GPT qui a été formée spécifiquement sur les données de l’entreprise ou de l’organisation qui l’utilise. Voir notre modèle GUM.
It's not just the public data on which the GPT model was originally built, but also the company's internal data - from documents reports, emails, meeting notes, and much more.
Cela signifie que cette version deGPT peut comprendre le jargon de l’entreprise, les produits et services spécifiques, les politiques internes, les processus, et même la culture d’entreprise. Il peut fournir des recommandations, des idées, et des informations plus pertinentes et spécifiques à l’entreprise que le modèle GPT standard.
Dans le cadre de l’innovation ouverte, ce GPT peut aider à faciliter et à améliorer la collaboration avec les parties externes. Il peut aider à la communication en traduisant le jargon de l’entreprise en termes plus généraux pour les partenaires externes. Il peut également aider à identifier les opportunités d’innovation en analysant les tendances du marché, les brevets, les publications académiques, et d’autres sources d’information externes.
De plus, le GPT étendu peut aider à la gestion de projets en fournissant des informations basées sur l’historique des projets de l’entreprise, en identifiant les risques potentiels basés sur des expériences passées, et en proposant des solutions basées sur des données internes et externes.
An extended TPM in the context of open innovation can act as a super collaborator who understands both the company's internal context and the wider external landscape, facilitating collaboration, innovation, and project success.
An example of GPT-assisted open innovation.
Suppose you're a project manager in an electronics manufacturing company. Your company has recently launched a project to improve the way it recycles electronic scrap. To achieve this, you are working with a mixed team consisting of an external chemist, an expert in solvent extraction of metals, and an in-house expert in environmental regulations.
From the outset of the project, the TPM, which has been trained on the company's data, proves invaluable. It provides relevant information to contextualize the project, drawing on internal research, previous recycling initiatives, regulatory implications and external case studies.
As the project progresses, the extended GPT facilitates communication between all team members. It helps to translate technical terms and complex concepts into a language that every team member can understand, while ensuring that everyone stays abreast of the latest advances and potential obstacles.
In addition, this TPM provides guidance based on the company's internal and external data. For example, it might indicate that the company has already tested a particular solvent which did not give the expected results. It could also share findings from other companies or research institutes on innovative metal extraction methods.
As the project progresses, the personalized generative AI adapts and learns from your interactions. If the team regularly encounters problems with one aspect of the project, the extended GPT could identify this pattern and propose solutions. It could also connect the team with other experts or resources if a particular need is identified.
The software layer associated with GPT.
Open innovation, supported by artificial intelligence technologies such as extended GPT, offers extraordinary potential for stimulating growth and creativity within organizations.
However, harnessing this potential effectively requires advanced platforms capable of precisely managing safety the contributions of each participant, while guaranteeing data confidentiality.
MARYLINK provides a platform for collaborative innovation, with intelligent workspaces that ensure granular access management. This enables different roles and individuals to specifically view and control the segments of information assigned to them.
By integrating a GPT trained on enterprise data, this platform enables innovative and secure collaboration. It optimizes the innovation process of organizations by guaranteeing precise control and robust protection of information.
Deutsch 
Español