diff --git a/doc/whats_new/v1.1.rst b/doc/whats_new/v1.1.rst
index b9651a1e1b6f8..329f87813a389 100644
--- a/doc/whats_new/v1.1.rst
+++ b/doc/whats_new/v1.1.rst
@@ -187,6 +187,11 @@ Changelog
   multilabel classification.
   :pr:`19689` by :user:`Guillaume Lemaitre <glemaitre>`.
 
+- |Enhancement| :class:`linear_model.RidgeCV` and
+  :class:`linear_model.RidgeClassifierCV` now raise consistent error message
+  when passed invalid values for `alphas`.
+  :pr:`21606` by :user:`Arturo Amor <ArturoAmorQ>`.
+
 - |Enhancement| :class:`linear_model.Ridge` and :class:`linear_model.RidgeClassifier`
   now raise consistent error message when passed invalid values for `alpha`,
   `max_iter` and `tol`.
diff --git a/sklearn/linear_model/_ridge.py b/sklearn/linear_model/_ridge.py
index 09ef46ffc7ebe..1426201a893aa 100644
--- a/sklearn/linear_model/_ridge.py
+++ b/sklearn/linear_model/_ridge.py
@@ -10,6 +10,7 @@
 
 
 from abc import ABCMeta, abstractmethod
+from functools import partial
 import warnings
 
 import numpy as np
@@ -1864,12 +1865,6 @@ def fit(self, X, y, sample_weight=None):
 
         self.alphas = np.asarray(self.alphas)
 
-        if np.any(self.alphas <= 0):
-            raise ValueError(
-                "alphas must be strictly positive. Got {} containing some "
-                "negative or null value instead.".format(self.alphas)
-            )
-
         X, y, X_offset, y_offset, X_scale = LinearModel._preprocess_data(
             X,
             y,
@@ -2038,9 +2033,30 @@ def fit(self, X, y, sample_weight=None):
         the validation score.
         """
         cv = self.cv
+
+        check_scalar_alpha = partial(
+            check_scalar,
+            target_type=numbers.Real,
+            min_val=0.0,
+            include_boundaries="neither",
+        )
+
+        if isinstance(self.alphas, (np.ndarray, list, tuple)):
+            n_alphas = 1 if np.ndim(self.alphas) == 0 else len(self.alphas)
+            if n_alphas != 1:
+                for index, alpha in enumerate(self.alphas):
+                    alpha = check_scalar_alpha(alpha, f"alphas[{index}]")
+            else:
+                self.alphas[0] = check_scalar_alpha(self.alphas[0], "alphas")
+        else:
+            # check for single non-iterable item
+            self.alphas = check_scalar_alpha(self.alphas, "alphas")
+
+        alphas = np.asarray(self.alphas)
+
         if cv is None:
             estimator = _RidgeGCV(
-                self.alphas,
+                alphas,
                 fit_intercept=self.fit_intercept,
                 normalize=self.normalize,
                 scoring=self.scoring,
@@ -2059,7 +2075,8 @@ def fit(self, X, y, sample_weight=None):
                 raise ValueError("cv!=None and store_cv_values=True are incompatible")
             if self.alpha_per_target:
                 raise ValueError("cv!=None and alpha_per_target=True are incompatible")
-            parameters = {"alpha": self.alphas}
+
+            parameters = {"alpha": alphas}
             solver = "sparse_cg" if sparse.issparse(X) else "auto"
             model = RidgeClassifier if is_classifier(self) else Ridge
             gs = GridSearchCV(
diff --git a/sklearn/linear_model/tests/test_ridge.py b/sklearn/linear_model/tests/test_ridge.py
index 1160e2db57fc6..975d16df06f12 100644
--- a/sklearn/linear_model/tests/test_ridge.py
+++ b/sklearn/linear_model/tests/test_ridge.py
@@ -1270,19 +1270,51 @@ def test_ridgecv_int_alphas():
     ridge.fit(X, y)
 
 
-def test_ridgecv_negative_alphas():
-    X = np.array([[-1.0, -1.0], [-1.0, 0], [-0.8, -1.0], [1.0, 1.0], [1.0, 0.0]])
-    y = [1, 1, 1, -1, -1]
+@pytest.mark.parametrize("Estimator", [RidgeCV, RidgeClassifierCV])
+@pytest.mark.parametrize(
+    "params, err_type, err_msg",
+    [
+        ({"alphas": (1, -1, -100)}, ValueError, r"alphas\[1\] == -1, must be > 0.0"),
+        (
+            {"alphas": (-0.1, -1.0, -10.0)},
+            ValueError,
+            r"alphas\[0\] == -0.1, must be > 0.0",
+        ),
+        (
+            {"alphas": (1, 1.0, "1")},
+            TypeError,
+            r"alphas\[2\] must be an instance of <class 'numbers.Real'>, not <class"
+            r" 'str'>",
+        ),
+    ],
+)
+def test_ridgecv_alphas_validation(Estimator, params, err_type, err_msg):
+    """Check the `alphas` validation in RidgeCV and RidgeClassifierCV."""
 
-    # Negative integers
-    ridge = RidgeCV(alphas=(-1, -10, -100))
-    with pytest.raises(ValueError, match="alphas must be strictly positive"):
-        ridge.fit(X, y)
+    n_samples, n_features = 5, 5
+    X = rng.randn(n_samples, n_features)
+    y = rng.randint(0, 2, n_samples)
 
-    # Negative floats
-    ridge = RidgeCV(alphas=(-0.1, -1.0, -10.0))
-    with pytest.raises(ValueError, match="alphas must be strictly positive"):
-        ridge.fit(X, y)
+    with pytest.raises(err_type, match=err_msg):
+        Estimator(**params).fit(X, y)
+
+
+@pytest.mark.parametrize("Estimator", [RidgeCV, RidgeClassifierCV])
+def test_ridgecv_alphas_scalar(Estimator):
+    """Check the case when `alphas` is a scalar.
+    This case was supported in the past when `alphas` where converted
+    into array in `__init__`.
+    We add this test to ensure backward compatibility.
+    """
+
+    n_samples, n_features = 5, 5
+    X = rng.randn(n_samples, n_features)
+    if Estimator is RidgeCV:
+        y = rng.randn(n_samples)
+    else:
+        y = rng.randint(0, 2, n_samples)
+
+    Estimator(alphas=1).fit(X, y)
 
 
 def test_raises_value_error_if_solver_not_supported():