Near Term (5-15 years): Early versions of quantum neural networks (QNNs) will start to use advanced quantum hardware with a few hundred to a thousand qubits. These systems will help improve tasks like recognizing digits and basic language understanding. New technology using tiny wires (nanowires) will show promise, allowing small groups of qubits to perform well. Some quantum methods will also begin to solve problems faster than traditional methods like TensorFlow for smaller tasks.

Mid Term (15-25 years): QNNs will grow and use systems with tens of thousands to hundreds of thousands of qubits. They will become much better at training AI models quickly, cutting down the time needed to learn. Operations on matrices (a way of organizing data) will happen almost instantly, which will improve how AI is trained and how it makes decisions. This will lead to significant progress in specific tasks like translating languages and analyzing feelings, by combining traditional and quantum methods seamlessly.