Microsoft CEO Satya Nadella emphasizes the necessity of human control over artificial intelligence and advocates for international cooperation to establish governance norms, particularly between the U.S. and China.
SEATTLE, WA—Microsoft Chairman and CEO Satya Nadella has asserted that artificial intelligence (AI) must remain under human control. He called for the establishment of international norms involving both governments and technology companies, highlighting the need for an understanding between the United States and China as a crucial starting point.
Speaking at the Ideas4India forum in Seattle on September 17, Nadella addressed the rapid advancements in AI capabilities while noting that significant questions regarding the control of this technology remain unresolved.
“We do not yet have the science of alignment, and that’s a fact,” he stated. “And yet we have powerful technology.”
Nadella emphasized that the guiding principle for AI development should be its value to humanity and the necessity for humans to maintain authority over the technology. He remarked, “Any technology that doesn’t serve humans or is not in control of humans is not worth doing.”
To illustrate the challenges faced by the AI industry, Nadella drew a parallel to a common dilemma in software development: when a critical flaw is identified, should the technology team address it immediately or allow the product to proceed? He suggested that the AI sector is grappling with a similar challenge, acknowledging that the forces shaping AI development make it unlikely to adopt an absolute stance on the issue. He stressed the importance of recognizing AI’s power and potential benefits while also striving to mitigate unintended consequences.
During an interactive session, Nadella reiterated the role of governments in AI governance, stating that the first requirement is to establish agreed norms between the United States and China. However, he cautioned that such dialogue should not exclude other nations. He urged countries to collaborate with AI developers to create shared norms that govern the technology.
Nadella also highlighted the significance of sovereignty at both the national and corporate levels. He pointed out that businesses continuously generate knowledge, which must remain under their control. “Countries need to have control of their knowledge,” he asserted.
This emphasis on knowledge protection ties into the broader theme of participation in AI development. Nadella argued that countries and companies must feel secure enough to engage fully in AI without being subjected to another entity’s technological advancements. “It will only happen when all of the world can both benefit from and contribute to this knowledge,” he said, referring to the development and dissemination of AI.
Additionally, Nadella cautioned against viewing observed improvements in AI as immutable scientific laws. He described scaling laws as empirical findings that have thus far held true, drawing a comparison to the historical trend known as Moore’s law.
As the conversation around AI continues to evolve, Nadella’s insights underscore the importance of maintaining human oversight and fostering international collaboration to ensure that the technology serves humanity effectively.
According to IANS, Nadella’s remarks at the Ideas4India forum reflect a growing consensus on the need for responsible AI governance.