Last Updated:
Many scientists working in leading AI labs have themselves described AI as a major threat. Anthropic researcher Evan Hubinger has predicted the extinction of the human race in the next decade. Now Microsoft CEO Satya Nadella has also made serious comments regarding AI.
Nadella said that the entire ecosystem should be responsible for AI security and direction.
New Delhi. Artificial Intelligence (AI) rapidly upgrading itself and going out of human control, this is the scary possibility about which top scientists and researchers around the world are worried. Jacob Coxon, a lead researcher who worked at OpenAI and Anthropic, has alleged that companies have become blind in the dangerous race for “self-improving superintelligence” without adequate security measures. Now Microsoft CEO Satya Nadella has warned about superintelligence. Nadella has bluntly said that if superintelligence cannot remain under human control and does not work for the good of humanity, then there is no justification in pursuing such technology.
While speaking on social media platform X, Satya Nadella said that the basic principle of development of superintelligence should be completely clear. He said that the first criterion for deciding the direction of AI should be human safety. However, Nadella did not demand a complete ban on the development of AI. He cautioned that companies should set a safe pace instead of rushing ahead without thinking. Additionally, robust evaluation mechanisms and wide participation must be ensured so that this technology does not remain a toy in the hands of a few. Satya Nadella said that Microsoft will issue a public code of conduct for its first-party MAI models and will take everyone’s suggestions on it.
Control over AI
Satya Nadella said that companies should maintain control over their “exclusive and tacit knowledge”. He cautioned that if companies hand over their organization’s knowledge and decision-making abilities to a single AI company, they run a huge risk of outsourcing their own thinking abilities. Nadella said that companies should not keep their data and infrastructure dependent on any one model, so that if needed, they can shift to different AI models without losing control.
Any pursuit of superintelligence has to be grounded in the core principle that if the AI we build is not helping humanity and under human control, it’s not worth pursuing.
We also need to accelerate and spread the benefits of AI, such that they are diffused widely across…
