Meta CEO Mark Zuckerberg recently published an op-ed in The Wall Street Journal pushing back against artificial intelligence “doomsday” narratives, making a core argument: superintelligence should not be monopolized by a few institutions but should become a tool available to everyone. He warned that the real danger is not AI itself, but the excessive concentration of power.<\/p>\n
In his July 28 article titled “The AI Future Is for Everyone,” Zuckerberg directly challenged the dominant narrative coming out of Silicon Valley. “The defining question of this era is not whether superintelligence will arrive, but who gets to use it,” he wrote. He criticized peers who peddle doomsday scenarios, arguing that the logic of “AI is so dangerous that only extreme concentration of power can keep us safe” is itself dangerous.<\/p>\n
Zuckerberg used the example of a superintelligent lawyer to illustrate the dangers of concentrated power: if only one person can afford a superintelligent lawyer, that person could dominate in court even when they’re in the wrong, making society worse off. But if everyone has access to a superintelligent lawyer, the justice system would actually become fairer and more efficient. He stressed that checks and balances \u2014 not reliance on “some benevolent superintelligence” \u2014 are the foundation of safety.<\/p>\n
These remarks come at a time when the AI industry is seeing a trend of “not releasing the most powerful models.” TIME has called this the new normal in AI: capabilities that are too powerful are held back from release. For example, Anthropic’s Claude Mythos and OpenAI’s GPT-Rosalind have not been made available to the public, restricted to vetted institutional clients only. Anthropic CEO Dario Amodei has even publicly stated that AI development should be slowed or paused when necessary.<\/p>\n
Zuckerberg argued that the more danger is emphasized, the more it leads to conclusions like “this can only be entrusted to a few” \u2014 and that is where the real danger lies. He contends that when everyone is empowered, people will naturally check each other’s power and collectively check the power of large institutions. He added that humanity is not a single culture, and values conflict with one another; any “single superintelligence” would inevitably sacrifice or violate the values of some group.<\/p>\n
Notably, OpenAI CEO Sam Altman expressed a similar view a few months ago: “Once you’ve seen AGI, you can’t unsee it.” He said the most dangerous impulse is “being the one who controls AGI,” and the only way out is “not letting anyone hold this magic ring alone.”<\/p>\n
In his article, Zuckerberg also discussed the future uses of AI. He believes early AI can only answer questions and handle routine tasks, but soon it will help people discover new knowledge \u2014 from developing new drugs to finding new ways to grow a business. On employment, his real concern is not AI automation itself, but the on-off switch for automation being in the hands of a very few. He emphasized balance: AI can be both an automation tool that replaces people and an empowerment tool that helps people expand their skills and start companies. He believes that as long as superintelligence is widely distributed, there will be more jobs in the future, not fewer.<\/p>\n
In his “Personal Superintelligence” manifesto released in 2025, Zuckerberg already made clear that in the face of new safety risks, Meta would “carefully decide which capabilities are appropriate to open source,” and that areas like biological risk require more coordination between governments and institutions.<\/p>\n
Meanwhile, OpenAI’s recently released “Building Abundant Intelligence” vision echoes similar ideas. The proposal argues that AI should evolve from a technology reserved for a few companies and experts into social infrastructure \u2014 like electricity, water systems, and communication networks \u2014 enabling “intelligence” itself to be supplied at scale, at low cost, and safely. OpenAI’s vision is for AI to be embedded in the foundation of all social activities, including education, healthcare, scientific research, and government services.<\/p>\n
This trend is triggering a massive wave of infrastructure investment globally. According to reports, tech giants including Microsoft, Google, Amazon, and Meta are shifting from pure AI model development competition to an “AI social infrastructure buildout race.” Data center construction investments on the order of trillions to tens of trillions of yen are underway around the world, covering cutting-edge GPUs, power systems, communications networks, and cooling equipment.<\/p>\n
For Japan’s industrial sector, this is seen as a major opportunity. TDK, Murata Manufacturing, and other major electronic component makers’ multilayer ceramic capacitors (MLCCs) and power components, Renesas Electronics’ power management semiconductors, and Sony Group’s imaging and optical technologies are all considered closely tied to AI infrastructure demand. The evolution of AI is becoming a new growth engine supporting Japan’s high-end manufacturing.<\/p>\n
Zuckerberg’s remarks will undoubtedly intensify the “open vs. closed” strategic debate in the AI industry. On one side, the “safety-first” camp represented by Anthropic advocates strict control over the most powerful AI systems. On the other, the “empowerment-first” camp represented by Meta pushes for open source and widespread distribution. The outcome of this debate will profoundly shape the future direction of AI technology and the competitive landscape of the global tech industry.<\/p>\n","protected":false},"excerpt":{"rendered":"Meta CEO Mark Zuckerberg recently published an op-ed in The Wall Street Journal pushing back against artificial intelligence…\n","protected":false},"author":2,"featured_media":175106,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":"","_share_on_mastodon":"0"},"categories":[154],"tags":[587,13286,586,79627,618,994,613,832,7212],"class_list":["post-175105","post","type-post","status-publish","format-standard","has-post-thumbnail","category-mark-zuckerberg","tag-anthropic","tag-claude-mythos","tag-dario-amodei","tag-gpt-rosalind","tag-mark-zuckerberg","tag-meta","tag-openai","tag-the-wall-street-journal","tag-time"],"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@people\/117024019158643194","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/posts\/175105","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/comments?post=175105"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/posts\/175105\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/media\/175106"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/media?parent=175105"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/categories?post=175105"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/people\/wp-json\/wp\/v2\/tags?post=175105"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}