Researchers have made a significant breakthrough in the field of quantum computing by investigating the trainability of photonic quantum circuits, a crucial component of variational quantum algorithms. The study focuses on passive linear-optical quantum circuits, which are prone to barren plateaus and high sampling costs, limiting their scalability. To address this, a novel framework is introduced, based on the ratio of sample variance to circuit variance, allowing for a more efficient training process. This framework enables the determination of the trainability of quantum circuits, paving the way for more robust and efficient quantum computing architectures. The findings have significant implications for the development of near-term quantum computing applications, particularly in the context of cryptography and computation. The advancement of quantum computing capabilities will ultimately challenge existing cryptographic systems, so practitioners must stay informed about these developments to ensure the long-term security of their systems1.