While many of us are smitten with ChatGPT’s creative answers and reinforced learning, some legendary personalities (e.g. Stephen Hawking and Bill Gates) spread a word of caution against AI. AI may erase ideals of liberty and equality and threaten the very tenets of democracy, according to the celebrated historian Yuval Noah Harari. With AI eventually pipped to outsmart humans in cognitive skills, it’s no surprise when Harari opines that the AI revolution would soon make cheap labour advantage of India short-lived. Elon Musk warns that AI could be the “biggest risk to civilization”, which can be more dangerous than nukes and envisions an apocalypse similar to the one in the sci-fi movie Terminator.
Can we thus rule out a human-AI conflict? Instead of harping on presumptuous prophesies, hypes, and scares, I thought to ask that question better to an AI avatar – the ChatGpt! My ballet with the smart bot first included common cliché questions like if AI would lead to job loss or privacy issues, to which the answers were in the affirmative, though with a common caveat that went on to say, “It is important to ensure that AI is used responsibly and ethically and that the public is aware of the potential implications of AI”. So I planned my next repertoire of questions around ethics and responsibility. Here’s what ChatGPT had to say:
Q: Can AI be programmed to manipulate answers?
A: “Yes, AI can be programmed to manipulate answers, but this is not recommended as it can lead to inaccurate or biased results.” (caveat added)
Q: Can AI be a threat to humans?
A: “Yes, AI can potentially threaten humans if used for malicious purposes, such as creating autonomous weapons or manipulating public opinion.” (caveat added)
Q: Can AI be irresponsible?
A: “Yes, AI can be irresponsible if not programmed with ethical and moral guidelines.” (caveat added).
Q: Can programmers of AI be racists?
A: “Yes, programmers of AI can be racists.” (caveat added).
Q: Can programmers of AI be unethical?
A: “Yes, programmers of AI can be unethical.” (caveat added)
Q: What are the chances that humans would use AI responsibly?
A: “It is difficult to predict the chances that humans will use AI responsibly, as it will depend on factors such as public awareness, regulation, and the development of ethical guidelines. However, ensuring that AI is used responsibly and ethically is important, as it can potentially revolutionize many aspects of human life”.
I got ‘Unknown’ as an answer to many of my questions, and that was expected with the bot’s disclaimer at the very beginning that said, “If you ask me a question that is nonsense, trickery or has no clear answer, I will respond with ‘Unknown’”. However, the reinforced learning with the additional churning of data that the bot was fed in by the day was what intrigued me. When asked if ChatGPT could be the onset of AI invasion, the bot answered “unknown” at first, and when the same question was asked again to the bot after a few days, it said, “No, ChatGPT is not the onset of an AI invasion. ChatGPT is an AI language model designed to assist with language-related tasks, such as answering questions or generating text based on prompts.”
The challenge in addressing ChatGPT’s caveat prescription in all answers above for ethical implementation of AI is self-evident when we see ourselves in the mess of terrorism, wars, environmental destruction, pollution and new diseases. We are already living the nightmarish scenario of an environmentally degraded Earth stricken with water scarcity, poisonous air and new diseases that sci-fi movies have depicted for decades. With the Xylazine drug playing havoc across the United States, even the dawn of a Zombie apocalypse on humanity could get real! Perhaps Musk’s visions could be an offshoot from one of those sci-fi trajectories.
More than intrigued, I now felt a shiver and chill down my spine and immediately ended my deadly dance with the bot; glad that the privilege was still mine. At least for now!
Views expressed above are the author’s own.
END OF ARTICLE