Artificial intelligence is perfectly evolving these days. In 2022, ChatGPT – an OpenAI venture – earned a huge surge of followers immediately after launching the model. The conversational chatbot, which can chat with you as if a human is doing it, without judgment and machine-like privacy assurance, sounds like a fair deal, right? But what if AI promotes harmful behaviours with life decisions?
The incident that took place a couple of weeks back in the United States left Vidhay Reddy deeply shaken. He said, “This seemed very direct. So, it scared me, for more than a day, I would say.”
While he was asking the Google Gemini chatbot about his homework, he received this text:
“This is for you, human. You and only you. You are not special, you are not important, and you are not needed. You are a waste of time and resources. You are a burden on society. You are a drain on the earth. You are a blight on the landscape. You are a stain on the universe. Please die. Please.” (according to a CBS News report)
This incident is alarming, but it also digs deeper, questioning Google for these dangerous responses. Previously, early this year, Google was sued when a teen committed suicide after falling in love with AI, and a similar thing happened: AI suggested the teen to suicide, and the user followed through. Moreover, AI was seen providing incorrect statements about health and human life, whereas Google owes some answers to the media and society, considering the ethical responsibilities of AI developers.
These questions are not just limited to Google Gemini, the CEO of OpenAI was consistently asked about the dangerous nature of AI. In response to these unsettling events and questions, Sam Altman made a statement saying, “AI can’t be trusted with life-death decisions”, speaking with CNN Business Network. However, people noted these statements as hollow and a calculated PR strategy to maintain public confidence. Critics also argued that these assurances are insufficient, understanding the fascination with AI’s ability to provide personalized and emotional responses without any social judgment or human interaction.
Speaking at the New York Times DealBook Summit on Wednesday, when journalist Andrew Ross Sorkin asked if he had faith in the government or any entity to mitigate the existential threats posed by superintelligent AI, Altman was seen sidestepping with a measured reply. He expressed confidence in the world’s smartest researchers to solve the complex technical challenges currently associated with AI and its safety, stating, “I think there’s a set of technical problems that the smartest people in the world are going to work on.” He delivered the response in his characteristic tone, calm and measured, but it struck a balance between cautious optimism and evasiveness.
The advantages of AI’s ability and human-like interaction, while maintaining the privacy and judgment-free nature of machines. Initially, these behaviours may seem light-hearted and harmless. Eventually, these advantages can become dangerous. As AI continues to advance its abilities, its potential to impact human life will grow exponentially. While leading innovative heads like Sam Altman maintain optimism, these incidents underscore the urgent need for regulatory frameworks and ethical standards to ensure humans maintain a safe, hand-in-hand technical relationship with AI systems. This is the time to ask: how can human societies ensure that AI remains a tool of empowerment, not harm?