diff --git a/.github/conda/meta.yaml b/.github/conda/meta.yaml
index a71fd602993..4b4a416d220 100644
--- a/.github/conda/meta.yaml
+++ b/.github/conda/meta.yaml
@@ -15,7 +15,7 @@ requirements:
     - python
     - pip
     - numpy >=1.17
-    - pyarrow >=15.0.0
+    - pyarrow >=16.0.0
     - python-xxhash
     - dill
     - pandas
@@ -30,7 +30,7 @@ requirements:
     - python
     - pip
     - numpy >=1.17
-    - pyarrow >=15.0.0
+    - pyarrow >=16.0.0
     - python-xxhash
     - dill
     - pandas
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index a4a9041fd28..3c6991b81ce 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -64,7 +64,7 @@ jobs:
         run: uv pip install --system --upgrade pyarrow huggingface-hub "dill<0.3.9"
       - name: Install dependencies (minimum versions)
         if: ${{ matrix.deps_versions != 'deps-latest' }}
-        run: uv pip install --system pyarrow==15.0.0 huggingface-hub==0.24.7 transformers dill==0.3.1.1
+        run: uv pip install --system pyarrow==16.0.0 huggingface-hub==0.24.7 transformers dill==0.3.1.1
       - name: Test with pytest
         run: |
           python -m pytest -rfExX -m ${{ matrix.test }} -n 2 --dist loadfile -sv ./tests/
diff --git a/setup.py b/setup.py
index 5e58064212a..d8566676eac 100644
--- a/setup.py
+++ b/setup.py
@@ -110,8 +110,8 @@
     # We use numpy>=1.17 to have np.random.Generator (Dataset shuffling)
     "numpy>=1.17",
     # Backend and serialization.
-    # Minimum 15.0.0 to be able to cast dictionary types to their underlying types
-    "pyarrow>=15.0.0",
+    # Minimum 16.0.0 to support string views
+    "pyarrow>=16.0.0",
     # For smart caching dataset processing
     "dill>=0.3.0,<0.3.9",  # tmp pin until dill has official support for determinism see https://github.com/uqfoundation/dill/issues/19
     # For performance gains with apache arrow
diff --git a/src/datasets/features/features.py b/src/datasets/features/features.py
index 70ea9fb5ddd..4fbd3ea7afd 100644
--- a/src/datasets/features/features.py
+++ b/src/datasets/features/features.py
@@ -110,6 +110,8 @@ def _arrow_to_datasets_dtype(arrow_type: pa.DataType) -> str:
         return "string"
     elif pyarrow.types.is_large_string(arrow_type):
         return "large_string"
+    elif pyarrow.types.is_string_view(arrow_type):
+        return "string_view"
     elif pyarrow.types.is_dictionary(arrow_type):
         return _arrow_to_datasets_dtype(arrow_type.value_type)
     else:
@@ -508,6 +510,7 @@ class Value:
     - `large_binary`
     - `string`
     - `large_string`
+    - `string_view`
 
     Args:
         dtype (`str`):
@@ -548,6 +551,10 @@ def encode_example(self, value):
             return float(value)
         elif pa.types.is_string(self.pa_type):
             return str(value)
+        elif pa.types.is_large_string(self.pa_type):
+            return str(value)
+        elif pa.types.is_string_view(self.pa_type):
+            return str(value)
         else:
             return value