Environment

• Qiskit version: main
• Python version: Any
• Operating system: Any

What is happening?

The docstring of the Qiskit UGate states that

Matrix representation:

.. math::

    \newcommand{\rotationangle}{\frac{\theta}{2}}

    U(\theta, \phi, \lambda) =
    \begin{pmatrix}
        \cos\left(\rotationangle\right) & -e^{i\lambda}\sin\left(\rotationangle\right) \\
        e^{i\phi}\sin\left(\rotationangle\right) & e^{i(\phi+\lambda)}\cos\left(\rotationangle\right)
    \end{pmatrix}

and

The matrix representation shown here is the same as in the OpenQASM 3.0 specification <https://openqasm.com/language/gates.html#built-in-gates>,
which differs from the OpenQASM 2.0 specification <https://doi.org/10.48550/arXiv.1707.03429> by a global phase of
:math:e^{i(\phi+\lambda)/2}.

which is not true when looking at the current OpenQASM 3 docs which state

\begin{split}U(\theta,\phi,\lambda) := \frac{1}{2}\left(\begin{array}{cc}
1+e^{i\theta} & -ie^{i\lambda}(1-e^{i\theta}) \
ie^{i\phi}(1-e^{i\theta}) & e^{i(\phi+\lambda)}(1+e^{i\theta}) \end{array}\right).\end{split}

How can we reproduce the issue?

Look at the docstrings.

What should happen?

There should be consistency in the definitions.

Any suggestions?

I suspect the note in the Qiskit docs shall be updated to reflect that this is not the OpenQASM3 U gate matrix and there is a global phase difference.