深潮TechFlow
深潮TechFlow|Jul 29, 2026 13:21
[Zuckerberg Criticizes AI Power Centralization, Supports Open Approach to Counter Anthropic and OpenAI] Deep Tide TechFlow News, July 29: According to a report by *The New York Times*, Meta CEO Mark Zuckerberg publicly criticized the closed AI development approach represented by Anthropic and OpenAI during an interview. He stated that their strict control over AI technology would lead to power centralization, limiting public access to this transformative technology, which he described as tantamount to 'betraying American technological values.' Zuckerberg expressed skepticism about the path of 'creating benevolent superintelligence through a single alignment mechanism' and advocated for providing 'personalized superintelligence' to cater to the diverse needs and values of different users. He also pointed out that the democratization of technology has always been the prevailing trend in the history of the tech industry. It is worth noting that Meta's own AI strategy has recently undergone changes. Last spring, the company invested $14.3 billion to acquire a stake in ScaleAI and brought its founder Alexandr Wang on board. Subsequently, Meta launched its first paid closed-source model, Muse Spark, while continuing to maintain its open-source model line. This statement aligns with NVIDIA CEO Jensen Huang's recent public support for open-source AI, with both parties standing in clear opposition to Anthropic and OpenAI's philosophies on AI development.
+5
Mentioned
Share To

Timeline

HotFlash

APP

X

Telegram

Facebook

Reddit

CopyLink

Hot Reads