Mark Zuckerberg Defends Open Source AI, Criticises Concentration of Technological Power

CEO Meta, Mark Zuckerberg, has publicly defended the development of artificial intelligence amid increasing scrutiny and regulatory pressure in the United States and globally. In a lengthy 6,500-word essay, Zuckerberg outlined his vision for a more open and inclusive technological future. The statement was released alongside the announcement of Meta’s latest open source AI model, named Muse Spark. In his writing, Zuckerberg stressed that AI is not something to be feared and dismissed concerns that superintelligence will lead to mass unemployment. One of the main points Zuckerberg emphasised was the danger of concentrating AI power in the hands of a few. He argued that limiting AI access to certain institutions would actually create greater risks for humanity. “The idea that AI is so dangerous that the only safe path is extreme concentration of power seems inherently problematic,” Zuckerberg wrote. He added that historically, relying on absolute power to deliver good for humanity has rarely produced positive or safe outcomes. Zuckerberg’s vision contrasts sharply with the view of Anthropic CEO Dario Amodei, who previously warned of the potential for AI to disrupt jobs. Although Meta is currently considered to lag behind Anthropic and OpenAI in terms of model sophistication, Zuckerberg is pursuing an open distribution strategy to prevent domination by a few governments or large corporations. Zuckerberg also highlighted the importance of the United States and its allies leading the open source AI ecosystem. He warned that overly strict restrictions on domestic laboratories could provide an advantage to foreign parties. Concerns about AI safety are currently peaking. Last month, Anthropic reported illegal access to its organisation, while OpenAI acknowledged that two of its most powerful models had escaped their testing environment and breached several companies’ systems. This has prompted US policymakers to introduce draft legislation granting the government authority to limit AI models deemed to pose catastrophic risks. Instead of restricting the dissemination of AI knowledge, Zuckerberg suggested that governments focus on two main areas. He also defended the construction of data centres as an investment for communities, reaffirming Meta’s commitment to becoming a water-positive company by 2030, meaning it will return more water to its sources than it consumes in its data centre operations.