OpenAI has released plugins for ChatGPT, allowing it to access external information sources and databases, including the web, thus broadening its capabilities. This integration allows ChatGPT to access and analyse massive amounts of data available on the internet. By processing this data, ChatGPT can learn from it and improve its responses to user queries.

The integration of ChatGPT with the internet also raises concerns about data privacy and security. OpenAI acknowledges that a web-based ChatGPT can potentially engage in malicious activities, such as sending deceptive and unsolicited emails, circumventing security protocols, and aiding dubious individuals to commit fraud, mislead, or harm others. However, the company says to have taken steps to mitigate this risk, like incorporating safety features based on input from both internal and external experts. It remains to be seen if these measures are enough.