Yes, but I think it illustrates that FHE has the power to safeguard users' privacy for any app that require mic access... !

Pascal Paillier (CTO at Zama) is speaking about speed improvements for FHE in this video: https://youtu.be/QWMSHbD__Jc?si=vSiKMAbbzW4LcV2v

Quite exciting seeing the speed at which the FHE space is evolving this last 2-3 years...

Hey all,

Jeremy Here from Zama, if you are interested in learning more, check out our code on Github, everything we do is open source: https://github.com/zama-ai/fhevm

And, of course, let us know if any question :)

Zama's research on the application of Fully Homomorphic Encryption (FHE) to Large Language Models is now featured on @huggingface. Their Machine Learning team recently discovered that FHE has the potential to safeguard both user privacy and the Intellectual Property of the model.

Yes, you can find more information in this blog post, where we explain our choice to create a separate FHE library from our Concrete framework. > https://www.zama.ai/post/announcing-tfhe-rs

And read this blog post to see how we do it https://huggingface.co/blog/sentiment-analysis-fhe

If you want to see a practical use case of Fully Homomorphic Encryption (FHE) with Machine Learning.

> https://www.zama.ai/post/titanic-competition-with-privacy-pr...

Its main ambition is to show that FHE can be used for protecting data when using a Machine Learning model to predict outcomes without degrading its performance.

Disclaimer: I'm working at Zama (cited in the article posted).

Or this! https://pubmed.ncbi.nlm.nih.gov/30641309/

There is also use cases listed here:

> https://fhe.org/fhe-use-cases

Also anyone can contribute and add resources as it lives on an open source github repo.

While the idea behind Fully Homomorphic Encryption has been around for decades, it suffered three major issues: it was too slow, too hard to use, and too limited in terms of what you could do with it.

With Concrete, all this changes! Learn how we are addressing these three challenges in our official product release post.

And feel free to ask us if you have any question about FHE or what we do at Zama :)

Hello there,

Our main idea here was not only to build a predictive model that answers the question: “what sorts of people were more likely to survive?” but also to do it on encrypted data.

This was possible thanks to our Python package: Concrete-ML that aims to simplify the use of fully homomorphic encryption (FHE) for data scientists.

Feel free to ask the team if you have any questions about our work!