How AI is uncovering new laws of physics
人工智慧如何揭示新的物理定律
Artificial Intelligence is no longer just a trend in tech; it is fundamentally changing how we understand the universe.
人工智慧已不再僅是科技圈的趨勢；它正從根本上改變我們對宇宙的認知。
In a major leap forward, AI is transitioning from simply confirming known laws of physics to uncovering entirely new ones.
在重大的飛躍中，人工智慧正從單純確認已知物理定律轉向發掘全新的定律。
A compelling example of this occurred in 2026, when researchers utilized specialized neural networks to study 'dusty plasma.'
2026年發生的一個引人注目的案例是，研究人員利用專門的神經網路來研究「塵埃電漿」。
By analyzing complex interactions in these systems, the AI identified new rules governing non-reciprocal forces that had previously eluded human scientists, achieving over 99% accuracy and challenging long-standing theoretical assumptions.
透過分析這些系統中複雜的相互作用，該人工智慧識別出了控制非互惠力的新規則，這些規則以往總是讓人類科學家難以捉摸，並達到了超過99%的準確率，從而挑戰了長期以來的理論假設。
The secret behind this success is 'symbolic regression,' a technique that allows AI to derive concise, interpretable equations from vast, messy datasets.
這項成功背後的秘密是「符號回歸」 (Symbolic regression)，這是一種允許人工智慧從龐大且凌亂的數據集中，推導出簡潔且具可解釋方程的技術。
Furthermore, by incorporating physical constraints like symmetry, these models can 'learn' physics without prior instructions.
此外，透過結合對稱性等物理約束，這些模型能夠在沒有預先指令的情況下「學習」物理學。
While some fear AI might replace scientists, the consensus is that this is a powerful partnership.
儘管有些人擔心人工智慧可能會取代科學家，但目前的共識是這是一種強而有力的夥伴關係。
