Qiskit QuantumKatas: Adapting Microsoft's Quantum Computing exercises for LLM evaluation

Researchers have successfully adapted Microsoft's QuantumKatas, a comprehensive quantum computing curriculum, to work with Qiskit, the most widely-used quantum computing framework. This adaptation enables the evaluation of large language models (LLMs) using a systematic framework, comprising 350 tasks across 26 categories. The benchmark covers a broad range of topics, from basic quantum gates to advanced algorithms, including Grover's, Simon's, and Deutsch-Jozsa algorithms. This development allows for a more nuanced understanding of LLMs' capabilities in quantum computing¹. The adaptation of QuantumKatas to Qiskit has significant implications for the field, as it shifts the focus from individual criminal activity to state-aligned efforts, requiring a fundamentally different approach to threat modeling. This matters to practitioners because it highlights the need for a new playbook in addressing quantum computing security threats, one that takes into account the complexities of geopolitical involvement.