Companies Are Tricking AI Chatbots, Should You Worry?

It seems that some companies are trying to play a sneaky game with the popular AI chatbots, like ChatGPT and Google AI Search. The game involves ‘stuffing’ websites like Reddit with their own promotional content, hoping the AI will pick it up and present it as fact.

Think of it like this: AI chatbots learn by reading vast amounts of information from the internet. If a company repeatedly posts positive things about its products on Reddit, even if it's not truly independent advice, the AI might see this information and then suggest those products to you when you ask for recommendations. This particular news story highlighted companies selling things like 'peptides' and 'hormone replacement therapies' using this tactic on a specific biohacking subreddit.

This isn't about the AI itself being bad; it's about people trying to trick the system. These chatbots are designed to give helpful summaries, but they can only work with the information they find. If that information is being quietly manipulated, it makes it harder for the AI to give you unbiased answers.

For Australian families and small businesses, this is a heads-up. When you use AI tools to research health products, services, or even local businesses, it’s a good reminder to always double-check the information. Just like you wouldn't trust every advertisement you see, it’s wise to be a little bit critical of what an AI tells you, especially if it seems to strongly favour one option. Always cross-reference with reliable, independent sources.