The artificial intelligence industry is currently changing. It is now focusing on creating 'superintelligence,' developing autonomous agents, and preparing major AI labs to go public on the stock market.

人工智慧產業目前正在轉型。現在的焦點在於創造「超智能」、開發自主代理，以及準備將大型 AI 實驗室推向股票市場上市。

Leading AI laboratories are changing how they operate to become more independent. For example, Microsoft AI, led by Mustafa Suleyman, has restructured to pursue superintelligence on its own. This reduces its reliance on OpenAI, although they still maintain a strategic partnership. Similarly, OpenAI and Anthropic have both filed private paperwork to launch initial public offerings (IPOs). These companies are moving from being research centers to full commercial businesses. Specifically, OpenAI wants to turn ChatGPT into a 'superapp' that includes coding tools and autonomous agents.

領先的 AI 實驗室正在改變其運作方式以變得更加獨立。例如，由 Mustafa Suleyman 領導的 Microsoft AI 已進行重組，以獨立追求超智能。這降低了對 OpenAI 的依賴，儘管雙方仍維持策略合作夥伴關係。同樣地，OpenAI 和 Anthropic 都已提交私人文件以啟動首次公開募股 (IPO)。這些公司正從研究中心轉型為全面的商業企業。具體而言，OpenAI 希望將 ChatGPT 轉變為一個包含程式碼工具和自主代理的「超級 App」。

Technologically, the industry is focused on 'recursive self-improvement' (RSI), where AI models help design and improve the next version of themselves. While some experts believe these claims are just a way to attract investors, the possibility of a rapid jump toward superintelligence has led to calls for international cooperation. Consequently, the United States and China would need to work together to agree on development pauses, despite their current political competition and the race for AI dominance.

在技術上，產業正專注於「遞歸自我完善」(RSI)，即 AI 模型協助設計並改善其下一個版本。雖然部分專家認為這些主張僅是吸引投資者的手段，但快速邁向超智能的可能性已導致國際合作的呼聲增加。因此，儘管美國與中國目前存在政治競爭且在 AI 主導權上激烈競爭，但兩國仍需合作商定開發暫停協議。

At the same time, AI use is splitting into two areas: business and consumers. In the business sector, companies are moving away from wasting computing power and are instead using 'intelligence allocation.' This means low-cost models handle simple tasks, while advanced models are used for complex reasoning. Meanwhile, consumers remain uncertain due to privacy concerns and accuracy issues. To address these challenges, governments are investing in their own infrastructure. For instance, the UK has invested £1.1 billion in domestic computing to reduce job losses caused by AI.

與此同時，AI 的應用正分化為兩個領域：商用與消費端。在商業領域，企業正擺脫對運算能力的浪費，轉而採用「智能分配」。這意味著低成本模型處理簡單任務，而高級模型則用於複雜推理。與此同時，消費者由於對隱私疑慮和準確性問題，仍持不確定態度。為了應對這些挑戰，各國政府正投資自有基礎設施。例如，英國已投資 11 億英鎊於國內運算，以減少 AI 造成的失業問題。

The AI sector is entering a phase of high-capital public markets and autonomous systems, while it continues to face serious challenges regarding global regulation and human safety.

AI 產業正進入一個高資本公開市場與自主系統的階段，同時在全球監管與人類安全方面仍面臨嚴重挑戰。