The news that former top researchers from Google DeepMind and OpenAI have secured billions of dollars in funding to develop so-called "Artificial Superintelligence" (ASI) is not merely a business success; it is the hallmark of a new, more ambitious, and perilous phase in technological history. As we move through the first half of 2026, the conversation has shifted from simple Large Language Models (LLMs) to the quest for an intelligence that surpasses human capabilities in every cognitive field.

The Brain Drain from Big Tech

Over the past two years, we have witnessed a massive exodus of talent from companies like Google and Meta. Researchers who were at the core of developing AlphaGo and Gemini are now choosing independence. The reason? A growing conviction that large corporations, bound by the need for immediate profits and shareholder pressure, cannot guarantee the safe development of superintelligence. These new entities, such as Safe Superintelligence (SSI) and similar initiatives, promise a "safety-first" approach, free from commercial distractions.

The mammoth funding, often reaching $10 billion in initial rounds, reflects investors' absolute faith in "scaling laws." The theory is simple yet expensive: the more compute and data you feed a model, the closer you get to Artificial General Intelligence (AGI). However, the cost of infrastructure in terms of GPUs and energy has turned AI development into a game for the few and the extremely powerful.

The Strategy of "Safe" Superintelligence

The question looming over the industry is whether "safety" is a genuine concern or clever marketing to attract capital and avoid regulatory scrutiny. The founders of these companies argue that ASI is inevitable and that building it in a controlled environment is the only way to avoid existential risks to humanity. According to analysts, this approach involves:

  • Market Isolation: Avoiding product releases for years so that research remains pure and focused.
  • Ethical Alignment: Developing systems that "understand" human values before they achieve full autonomy.
  • Energy Autonomy: Investing in nuclear energy and proprietary data centers to support immense computational needs.

Impact on Global Economy and Geopolitics

This concentration of capital around a few individuals creates new inequalities. While Europe attempts to regulate AI through the AI Act, Silicon Valley and China are engaged in an arms race. Funding for these startups no longer comes just from venture capitals but also from sovereign wealth funds that see ASI as the ultimate geopolitical advantage. Whoever controls superintelligence will control the global economy, medical research, and even military strategy.

"We are not just building a tool; we are building the last inventor humanity will ever need," one of the protagonists of this new race recently stated.

The challenge for society remains: How can we ensure that the benefits of such technology are distributed to everyone and do not remain in the hands of an elite group of researchers and investors? The history of technology has taught us that the concentration of power rarely leads to democratization without strict oversight.

Conclusion: A Leap into the Unknown

The mammoth funding we see today is the bet of the millennium. If these researchers succeed, our world will change radically within the next decade. If they fail, we will be talking about the largest financial bubble in history. What is certain is that the path to superintelligence is irreversible, and the players at the table have now bet everything.