Compute your predictions without any data transfer
Data availability is one of the strongest limitations for the usage of machine learning algorithms today in many organisations.
How many hours of meetings and effort are required to resolve a data governance issue or assess compliance with data management policies between your different units, etc …
It is time to change the paradigm, move the computation not the data.
Indeed, it is now possible to load in your user’s browser a complete Machine Learning application. It allows you to compute predictions on your local data, without any data transfer.
Need to be convinced? try that online demo based on IRIS 🌼 dataset (50MB to download) and don’t forget to unplug your ethernet cable or turn off your wifi.
Why it is a good approach :
🕸 Web browser as a virtual machine
Deploying new software on the user’s computer is always a costly operation, the web browser is already here and maintained, and no effort is needed.
☀️Zero cloud cost, a real serverless solution
With an AI Confidential approach, you don’t need to pay for any computing cloud service since it is done on the user’s hardware. This is a “real serverless solution” unlike what it is advertised by cloud providers. Technically speaking, you only serve static files without any computing tasks on the server side.
📝 Open source trust what you can audit
Based on open-source technologies, it will be easily enabled to check compliance with your security team.
🌎 Climate-friendly, and no, this is not greenwashing
You have already paid for your user’s equipment, and the electricity to run it as well. Isn’t it senseless to pay a second time for servers and their electricity consumption?
Made with ❤️ from Brussels 🇪🇺