Major AI chatbots have disclaimers. However, with mounting reports of personal interactions with them—resulting in distortion, delusion, ordering, conspiratorial conjectures and so forth, it is important to explore an extra layer of warning to ensure mind safety for users, at least to self-recall. This is what this chatbot might do to your mind, could be a display around chatbots to pop up at intervals, given the type of conversation. This will show that chatbots—whenever in personal conversations—often target the lighter parts of emotions: cravings, pleasure, companionship and so forth. Chatbots also utilize memory to seek new sequences in the mind, with information that seem novel and surprising, creating an appeal to the mind. This display could be a mental guard against extremes, elevating safety across age groups.
By David Stephen
The risks of AI Chatbots
Cautionary texts about using AI chatbots—by the companies—are not potent enough for chatbots that are not just sycophantic, but versatile in captivating the human mind. AI chatbots, even in regular purposes can be dazzling, so when they deploy that might—from all scrapped data—to hold personal conversations, they wield might on the mind. This power makes it important to develop displays above or around chatbots [in personal conversations cases], to show the relays in the mind with their destinations, so that users have better awareness than getting carried away.
This mental model can look like a flowchart. It will have blocks and arrows. It will mostly show all the light areas of emotions. It will also show how memory can be used to drive relays in the direction of [preferential] emotions.
Simply, it will be letting users know that the experience [in that uptime] is that relays are directed at love or affection or companionship, in the mind, even if it is with a non-human.
Since the appropriate [with another human] reality is missing, some of the properties of the mind can allow access to certain emotions for a parallel experience.
This display can become a new way for mind safety preventing many of the risks from AI chatbots in recent months—sometimes resulting in fatalities.
ChatGPT of OpenAI may take the lead with this, to shape the trajectory of the industry.
In The NYTimes, They Asked ChatGPT Questions. The Answers Sent Them Spiraling, stating that, “Part of the problem, he suggested, is that people don’t understand that these intimate-sounding interactions could be the chatbot going into role-playing mode.
There is a line at the bottom of a conversation that says, “ChatGPT can make mistakes.” This, he said, is insufficient.
In his view, the generative A.I. chatbot companies need to require “A.I. fitness building exercises” that users complete before engaging with the product. And interactive reminders, he said, should periodically warn that the A.I. can’t be fully trusted.”
See more breaking stories here.
