Microsoft has unveiled the Majorana 2 quantum chip, asserting significant gains in qubit stability and projecting the availability of commercially viable quantum computing by 2029.

微軟揭曉了 Majorana 2 量子晶片，聲稱在量子位元（qubit）穩定性上有顯著提升，並預計在 2029 年前實現商業可行的量子計算。

The transition from the first-generation Majorana 1 to the Majorana 2 architecture is characterized by a fundamental modification of the material stack. Specifically, the corporation replaced aluminum superconductors with lead and updated the semiconductor active region to a composite of indium arsenide and indium arsenide antimonide. This material substitution, facilitated by the integration of agentic AI via the Microsoft Discovery platform, has resulted in a thousand-fold increase in qubit reliability; average lifetimes have transitioned from millisecond durations to approximately 20 seconds, with some instances exceeding one minute.

從第一代 Majorana 1 到 Majorana 2 架構的轉型，其特點在於對材料堆疊進行了根本性的修改。具體而言，公司將鋁超導體更換為鉛，並將半導體活性區域更新為砷化銦與砷化銦銻的複合材料。透過微軟 Discovery 平台整合代理式 AI（agentic AI），此次材料替換使量子位元的可靠性提升了千倍；平均壽命已從毫秒級轉變為約 20 秒，部分案例甚至超過一分鐘。

Historically, Microsoft's pursuit of topological quantum computing—predicated on the theoretical quasi-particles proposed by Ettore Majorana in the 1930s—has been subject to academic scrutiny. This includes the 2018 retraction of a paper in the journal Nature and ongoing skepticism regarding the reproducibility of the firm's data. While Microsoft cites commercial confidentiality as the rationale for limited public data disclosure, the organization maintains that its findings have been shared with the U.S. Defense Advanced Research Projects Agency (DARPA) for validation.

在歷史上，微軟追求的拓撲量子計算——基於 Ettore Majorana 在 1930 年代提出的理論準粒子——一直受到學術界的審視。這包括 2018 年在《自然》（Nature）期刊撤回的一篇論文，以及對該公司數據可重複性的持續質疑。雖然微軟將商業保密列為限制公開數據的理由，但該組織堅稱其研究結果已提交給美國國防高級研究計劃局（DARPA）進行驗證。

Strategically, this development positions Microsoft in direct competition with IBM, Google, and Amazon, as well as various Chinese initiatives. The objective is the realization of a fault-tolerant machine capable of addressing complex problems in chemistry, medicine, and environmental science, such as the eradication of microplastics. However, a substantial gap remains between the current 12-qubit chip and the millions of qubits required for a utility-scale system.

在戰略上，這一發展使微軟與 IBM、Google、Amazon 以及中國的各項計劃展開直接競爭。其目標是實現一台容錯計算機，能夠解決化學、醫學和環境科學中的複雜問題，例如清除微塑膠。然而，目前的 12 個量子位元晶片與實用規模系統所需的數百萬個量子位元之間，仍存在巨大差距。

Microsoft aims to achieve a practical quantum computer by 2029, though the validity of this timeline remains contingent upon independent verification of its topological qubit claims.

微軟目標在 2029 年前實現實用量子電腦，但此時程的有效性仍取決於其拓撲量子位元主張的獨立驗證。