diff --git a/featuretools/primitives/standard/transform/natural_language/__init__.py b/featuretools/primitives/standard/transform/natural_language/__init__.py
index bfa389e9df..2d66c1c06a 100644
--- a/featuretools/primitives/standard/transform/natural_language/__init__.py
+++ b/featuretools/primitives/standard/transform/natural_language/__init__.py
@@ -1,3 +1,6 @@
+from featuretools.primitives.standard.transform.natural_language.openai import (
+    OpenAIEmbeddings,
+)
 from featuretools.primitives.standard.transform.natural_language.count_string import (
     CountString,
 )
diff --git a/featuretools/primitives/standard/transform/natural_language/openai/__init__.py b/featuretools/primitives/standard/transform/natural_language/openai/__init__.py
new file mode 100644
index 0000000000..fa0155f804
--- /dev/null
+++ b/featuretools/primitives/standard/transform/natural_language/openai/__init__.py
@@ -0,0 +1,3 @@
+from featuretools.primitives.standard.transform.natural_language.openai.embeddings import (
+    OpenAIEmbeddings,
+)
diff --git a/featuretools/primitives/standard/transform/natural_language/openai/embeddings.py b/featuretools/primitives/standard/transform/natural_language/openai/embeddings.py
new file mode 100644
index 0000000000..dc352c50a3
--- /dev/null
+++ b/featuretools/primitives/standard/transform/natural_language/openai/embeddings.py
@@ -0,0 +1,65 @@
+import numpy as np
+import pandas as pd
+import tiktoken
+from openai.embeddings_utils import get_embedding
+from woodwork.column_schema import ColumnSchema
+from woodwork.logical_types import Double, NaturalLanguage
+
+from featuretools.primitives.base import TransformPrimitive
+from featuretools.primitives.standard.transform.natural_language.openai.model import (
+    OpenAIEmbeddingModel,
+)
+
+DEFAULT_MODEL = OpenAIEmbeddingModel(
+    name="text-embedding-ada-002",
+    encoding="cl100k_base",
+    max_tokens=8191,
+    output_dimensions=1536,
+)
+
+
+class OpenAIEmbeddings(TransformPrimitive):
+    """Generates embeddings using OpenAI.
+
+    Description:
+        Given list of strings, determine the embeddings for each string, using
+        the OpenAI model.
+
+    Args:
+        model (OpenAIEmbeddingModel, optional): The model to use to produce embeddings.
+            Defaults to "text-embedding-ada-002" if not specified.
+
+    Examples:
+        >>> x = ['This is a test file', 'This is second line', 'third line $1,000', None]
+        >>> openai_embeddings = OpenAIEmbeddings()
+        >>> openai_embeddings(x).tolist()
+        [4.0, 4.0, 5.0, nan]
+    """
+
+    name = "openai_embeddings"
+    input_types = [ColumnSchema(logical_type=NaturalLanguage)]
+    return_type = ColumnSchema(logical_type=Double, semantic_tags={"numeric"})
+
+    def __init__(self, model=DEFAULT_MODEL):
+        self.model = model
+        self.number_output_features = model.output_dimensions
+
+    def get_function(self):
+        encoding = tiktoken.get_encoding(self.model.encoding)
+
+        def is_too_many_tokens(element):
+            return len(encoding.encode(element)) > self.model.max_tokens
+
+        def get_openai_embeddings(series):
+            invalid = [np.nan] * self.number_output_features
+            result = []
+            for element in series:
+                if pd.isnull(element) or is_too_many_tokens(element):
+                    result.append(invalid)
+                else:
+                    embedding = get_embedding(element, engine=self.model.name)
+                    result.append(embedding)
+            result = np.array(result).T.tolist()
+            return pd.Series(result)
+
+        return get_openai_embeddings
diff --git a/featuretools/primitives/standard/transform/natural_language/openai/model.py b/featuretools/primitives/standard/transform/natural_language/openai/model.py
new file mode 100644
index 0000000000..5d20b8732a
--- /dev/null
+++ b/featuretools/primitives/standard/transform/natural_language/openai/model.py
@@ -0,0 +1,16 @@
+class OpenAIModel(object):
+    """A model accessible via the OpenAI API."""
+
+    def __init__(self, name, encoding, max_tokens):
+        self.name = name
+        self.encoding = encoding
+        self.max_tokens = max_tokens
+        pass
+
+
+class OpenAIEmbeddingModel(OpenAIModel):
+    """A model accessible via the OpenAI API that can produce embeddings."""
+
+    def __init__(self, name, encoding, max_tokens, output_dimensions):
+        self.output_dimensions = output_dimensions
+        super().__init__(name, encoding, max_tokens)
diff --git a/pyproject.toml b/pyproject.toml
index b5634871fa..5a2b941bef 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,6 +40,8 @@ dependencies = [
     "scipy >= 1.4.0",
     "tqdm >= 4.32.0",
     "woodwork[dask] >= 0.18.0",
+    "openai[embeddings] >= 0.26.5",
+    "tiktoken >= 0.3.0",
 ]
 
 [project.urls]