Risks and Limitations of ChatGPT

ChatGPT has several risks and limitations, including:

• The tendency to create false information
• Its potential to create deceptive content
• Potential bias in its training data
• Security risks from the data it is gathering

ChatGPT often produces misinformation, or information that isn’t true. Its tendency to make up false facts is part of a problem known as AI hallucination.

ChatGPT has the potential to produce disinformation, or information meant to deceive others. People can use prompts to spread propaganda or misleading content.

ChatGPT is collecting data on the ways it is being used, including information being put into prompts. This is important to consider when entering sensitive information.

ChatGPT will sometimes reproduce biases found in its training data. These biases have the potential to produce inappropriate or harmful content.