# True-AI Companies

This week I had the opportunity to get to know a true-AI startup, [**Jina AI**](https://www.linkedin.com/feed/#), and its experienced ML engineer, [**Isabelle Mohr**](https://www.linkedin.com/in/isabelle-mohr-a8ba5516b/), during the **#GenerativeAI** Meetup organized by [**Roosh Circle**](https://www.linkedin.com/company/roosh-circle/) and hosted by [**SAP**](https://www.sap.com/) in [**SAP Data Space**](https://www.sap.com/germany/about/berlin/data-space.html).

Why call them a true-AI startup? Many so-called AI companies exist, but a true AI company—in my opinion—must have a team of AI/ML/data scientists who are actively pushing the boundaries, which is completely true about Jina AI.

The primary product of Jina is an **#open\_sourced** embedding model that has outperformed the competition despite being more **#sustainable** (smaller model size and less training cost). Although it is a continuous competition with ups and downs, they are clearly showcasing how to be more sustainable in AI while being more reliable.

![](https://cdn.hashnode.com/res/hashnode/image/upload/v1717856030361/48c5af3a-14c8-4935-b5fc-9bc9cd78d694.jpeg align="left")

Another highlight from the event was the presentation from our AI Development Expert, [**Mathis Börner**](https://www.linkedin.com/in/mathisboerner/). In it, he provided a nice overview of **#SAP** Gen AI projects, showing how we are comprehensively playing a role in the field to provide **#relevant**, **#reliable**, and **#responsible** AI-powered services while preparing for future innovations.

Hope all the best for Jina AI and all true-AI companies which are shapping a sustainable and ethical AI-powered future.

## Plus

I learned about this interesting paper from Isabelle and hopfully will share a summary when I can finish reading it.

*   [Train Short, Test Long: Attention with linear biases enables input length extrapolation](https://arxiv.org/pdf/2108.12409)