Bad Qubits — Learn Quantum by Doing

Multiple choice

The Gottesman-Knill theorem concerns circuits using only computational-basis state preparation, Clifford gates (H, S, CNOT), and computational-basis measurement. What does the theorem establish about such circuits?

They can be simulated on a classical computer in time polynomial in the number of qubits and gates
They are universal for quantum computation on their own, with no extra gates needed
They require tracking all 2^n amplitudes and so are exponentially hard to simulate classically
They cannot create any entanglement between qubits