The billionaire entrepreneur and CEO of SpaceX and Tesla, Elon Musk, has claimed to be developing a new artificial intelligence (AI) system dubbed “TruthGPT.” During an interview with Fox News’ Tucker Carlson, Musk characterized TruthGPT as a “maximum truth-seeking AI”. It stated that it is being developed as a safer alternative to ChatGPT, the AI language model developed by OpenAI.

According to Musk, TruthGPT is intended to focus on comprehending the universe, as opposed to simply learning from human interactions. He believes that an artificial intelligence that is sincerely interested in comprehending the world around it is less likely to threaten humanity than one that is only concerned with maximizing its own objectives.

Musk has been a vocal critic of OpenAI, the non-profit AI software organization he co-founded in 2015. Musk has since criticized the organization for prioritizing commercial interests over its stated mission, despite the fact that it was founded to advance AI for the benefit of all humanity. Musk framed TruthGPT as a course correction for OpenAI in his interview with Carlson, which he believes has lost its way.

TruthGPT: Designed to be a safer alternative to OpenAI’s ChatGPT

The concept of an AI system that pursues the truth instead of simply optimizing results is not novel. It is a fundamental principle of numerous ethical AI frameworks, such as the IEEE Global Initiative on Ethics of Autonomous and Intelligent Systems. Such frameworks aim to ensure that artificial intelligence is designed and developed in accordance with human values and human rights.

However, the concept of a “maximum truth-seeking AI” poses a number of intriguing concerns. How would such an AI, for instance, prioritize competing truths? What criteria would it employ to assess the veracity of various claims? How would it reconcile its pursuit of the truth with other factors such as privacy and security?

Despite these obstacles, the concept of an AI system designed to prioritize truth and comprehension is undeniably alluring. If successful, TruthGPT could help resolve some of the concerns raised about the potential dangers of artificial intelligence. It could also pave the way for a new era of artificial intelligence development centered on human values and ethics.