Concerned that ChatGPT may Record your Data? Learn about PrivateGPT

This open-source architecture, PrivateGPT, operates locally on your device and may be tweaked to offer solutions pertinent to files

Preeti Anand
New Update

ChatGPT is a conversational interaction model that can respond to follow-up queries, acknowledge mistakes, refute false premises, and reject unsuitable requests. An excellent AI product, ChatGPT has countless uses and continually opens up new opportunities. But there are some issues with it. According to its privacy statement, the chatbot may record any input you give it. Several companies, including Samsung, JPMorgan, and Apple, have forbidden their staff from using ChatGPT out of concern that sensitive data may be exposed.


You can attempt an alternative if you share my concerns regarding personal information. This open-source architecture, PrivateGPT, operates locally on your device and may be tweaked to offer solutions pertinent to your files and documents. The chatbot can function entirely offline and doesn't need an internet connection, making it even better.

How To Use PrivateGPT?

The user must first download the chatbot on their computer to utilise it. Iván Martnez Toro developed the chatbot. Users are then told to copy their pertinent files into a folder so the chatbot can import all the information. No data leaves your local environment during the ingest process, according to the PrivateGPT guidelines, which may even be done without an internet connection.


You can ask your queries in a Python execution environment

You can ask questions in a Python execution environment once the setup process is finished. The processing time for each prompt is roughly 20 to 30 seconds, which is slower than ChatGPT and Google Bard, but on the other hand, your prompts are processed fully on your system rather than on powerful supercomputers. When finished, the chatbot should have displayed your response and the context from 4 sources in your papers.

According to Toro, who invented PrivateGPT after learning how helpful ChatGPT is in the workplace, the AI chatbot is currently just a proof-of-concept to explore if it is feasible to construct completely localised chatbots similar to ChatGPT.



As AI chatbots like ChatGPT become more capable and adaptable, privacy concerns for businesses using them in various industries increasingly arise. Some companies are considering developing large language models (LLMs) that may be customised for their internal papers to protect sensitive data. They can provide customised work support to their staff in this way. One significant illustration of this is Samsung, which is said to have collaborated with Naver to create its own proprietary ChatGPT-like AI platform.