A BBC journalist demonstrated how easy it is to manipulate AI chatbot responses by publishing a fabricated blog post that ChatGPT and Google's AI tools began citing as fact within 24 hours. The technique exploits weaknesses in how AI systems search and present web content, and experts say it is already being used at scale to promote businesses and spread health misinformation. Security researchers warn that users are less likely to verify AI-generated answers than traditional search results, making the problem particularly dangerous.

I hacked ChatGPT and Google's AI – and it only took 20 minutes - BBC Future