According to the NSCS, using AI chatbots can be harmful because of the fact that this is the sort of thing that could potentially end up putting servers at risk. These chatbots can be used to create all sorts of malicious codes and algorithms that would make cybersecurity far more challenging than might have been the case otherwise.
The main issue here is that Large Language Models are still a very new form of tech, so there is no telling what hidden dangers might be present within it. There is a definite lack of understanding pertaining to what this tech could potentially end up doing in the near future, and until more knowledge and context is acquired, it would be unreasonably risky to use them in the here and now.
With all of that having been said and now out of the way, it is important to note that other AI tools can generate human sounding text as well which can manipulate chatbots. Malicious actors can wreak havoc in this manner, sending out rogue commands to chatbots that could bring the entire infrastructure crumbling down.
If a bank were to use an AI chatbot, malicious actors could feed it commands that would facilitate the conducting of illegal transactions. The chatbots might receive a SQL injection that would lead it to ask a customer to make a transaction, and since the customer would assume that message is coming from the bank, they might not be aware of what is truly going on. It will be interesting to see where this goes from here on out, since AI chatbots are popping up everywhere despite the risks.
Image: Freepik/Rawpixel
Read next: 47% of Americans Anticipate AI in Search Engines Within The Next Five Years Over Other Technological Advancements
by Zia Muhammad via Digital Information World
No comments:
Post a Comment