diff --git a/.github/workflows/unit-tests.yml b/.github/workflows/unit-tests.yml
index dd5d090e098b0..a3cffb4b03b93 100644
--- a/.github/workflows/unit-tests.yml
+++ b/.github/workflows/unit-tests.yml
@@ -92,7 +92,10 @@ jobs:
           - name: "Numpy Dev"
             env_file: actions-311-numpydev.yaml
             pattern: "not slow and not network and not single_cpu"
-            test_args: "-W error::DeprecationWarning -W error::FutureWarning"
+            # Currently restricted the warnings that error to Deprecation Warnings from numpy
+            # done since pyarrow isn't compatible with numpydev always
+            # TODO: work with pyarrow to revert this?
+            test_args: "-W error::DeprecationWarning:numpy -W error::FutureWarning:numpy"
           - name: "Pyarrow Nightly"
             env_file: actions-311-pyarrownightly.yaml
             pattern: "not slow and not network and not single_cpu"
diff --git a/pandas/__init__.py b/pandas/__init__.py
index 7fab662ed2de4..ed524c2bb3619 100644
--- a/pandas/__init__.py
+++ b/pandas/__init__.py
@@ -202,8 +202,37 @@
         FutureWarning,
         stacklevel=2,
     )
-# Don't allow users to use pandas.os or pandas.warnings
-del os, warnings
+
+# DeprecationWarning for missing pyarrow
+from pandas.compat.pyarrow import pa_version_under10p1, pa_not_found
+
+if pa_version_under10p1:
+    # pyarrow is either too old or nonexistent, warn
+    from pandas.compat._optional import VERSIONS
+
+    if pa_not_found:
+        pa_msg = "was not found to be installed on your system."
+    else:
+        pa_msg = (
+            f"was too old on your system - pyarrow {VERSIONS['pyarrow']} "
+            "is the current minimum supported version as of this release."
+        )
+
+    warnings.warn(
+        f"""
+Pyarrow will become a required dependency of pandas in the next major release of pandas (pandas 3.0),
+(to allow more performant data types, such as the Arrow string type, and better interoperability with other libraries)
+but {pa_msg}
+If this would cause problems for you,
+please provide us feedback at https://github.com/pandas-dev/pandas/issues/54466
+        """,  # noqa: E501
+        DeprecationWarning,
+        stacklevel=2,
+    )
+    del VERSIONS, pa_msg
+
+# Delete all unnecessary imported modules
+del pa_version_under10p1, pa_not_found, warnings, os
 
 # module level doc-string
 __doc__ = """
diff --git a/pandas/compat/pyarrow.py b/pandas/compat/pyarrow.py
index beb4814914101..2e151123ef2c9 100644
--- a/pandas/compat/pyarrow.py
+++ b/pandas/compat/pyarrow.py
@@ -8,6 +8,7 @@
     import pyarrow as pa
 
     _palv = Version(Version(pa.__version__).base_version)
+    pa_not_found = False
     pa_version_under10p1 = _palv < Version("10.0.1")
     pa_version_under11p0 = _palv < Version("11.0.0")
     pa_version_under12p0 = _palv < Version("12.0.0")
@@ -16,6 +17,7 @@
     pa_version_under14p1 = _palv < Version("14.0.1")
     pa_version_under15p0 = _palv < Version("15.0.0")
 except ImportError:
+    pa_not_found = True
     pa_version_under10p1 = True
     pa_version_under11p0 = True
     pa_version_under12p0 = True
diff --git a/pandas/tests/test_common.py b/pandas/tests/test_common.py
index e8a1c961c8cb6..fe24755e8cc23 100644
--- a/pandas/tests/test_common.py
+++ b/pandas/tests/test_common.py
@@ -8,6 +8,8 @@
 import numpy as np
 import pytest
 
+import pandas.util._test_decorators as td
+
 import pandas as pd
 from pandas import Series
 import pandas._testing as tm
@@ -265,3 +267,23 @@ def test_bz2_missing_import():
     code = textwrap.dedent(code)
     call = [sys.executable, "-c", code]
     subprocess.check_output(call)
+
+
+@td.skip_if_installed("pyarrow")
+@pytest.mark.parametrize("module", ["pandas", "pandas.arrays"])
+def test_pyarrow_missing_warn(module):
+    # GH56896
+    response = subprocess.run(
+        [sys.executable, "-c", f"import {module}"],
+        capture_output=True,
+        check=True,
+    )
+    msg = """
+Pyarrow will become a required dependency of pandas in the next major release of pandas (pandas 3.0),
+(to allow more performant data types, such as the Arrow string type, and better interoperability with other libraries)
+but was not found to be installed on your system.
+If this would cause problems for you,
+please provide us feedback at https://github.com/pandas-dev/pandas/issues/54466
+"""  # noqa: E501
+    stderr_msg = response.stderr.decode("utf-8")
+    assert msg in stderr_msg, stderr_msg