Add insurance AI use case examples by oludarebakare · Pull Request #1056 · microsoft/generative-ai-for-beginners

oludarebakare · 2025-12-13T00:02:31Z

This pull request adds a new markdown file, insurance-ai-usecases.md, which outlines practical generative AI applications for Nigerian insurance companies. Use cases include claims processing, fraud detection, customer support, policy simplification, and regulatory compliance.

github-actions · 2025-12-13T00:02:46Z

👋 Thanks for contributing @oludarebakare! We will review the pull request and get back to you soon.

oludarebakare · 2025-12-13T00:07:17Z

@microsoft-github-policy-service agree

koreyspace · 2026-01-12T10:43:40Z

Hi @oludarebakare thanks for the contribution here but I dont think this should be a seperate files within this course. If you would liek to add content to existing lessons to explain the use cases further, I think that makes that will be helpful.

Copilot

Pull request overview

This pull request adds documentation and a demo script about generative AI applications for Nigerian insurance companies. However, the implementation has critical issues: the demo code doesn't actually use any AI/GenAI functionality, contradicting the repository's educational focus on Generative AI applications.

Changes:

Added insurance-ai-usecases.md documenting five AI use cases for insurance companies (claims processing, fraud detection, customer support, policy simplification, regulatory compliance)
Added claims-ai-demo.py with a basic text summarization function and example claim data

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 4 comments.

File	Description
insurance-ai-usecases.md	Markdown documentation describing practical generative AI use cases for Nigerian insurance companies
claims-ai-demo.py	Python script demonstrating claims summarization (though currently using only string manipulation, not actual AI)

Comments suppressed due to low confidence (2)

claims-ai-demo.py:18

The output label "AI-Generated Summary" is misleading since the summarize_claim function doesn't use any AI or machine learning - it's just basic string manipulation. Update this label to accurately reflect what the code does, or better yet, implement actual AI-based summarization.

print("\nAI-Generated Summary:")

insurance-ai-usecases.md:20

The hyphen in "customer‑friendly" is a non-breaking hyphen (Unicode U+2011) instead of a standard hyphen. This should be replaced with a regular hyphen-minus character for consistency.

- Example: Explaining NDPR compliance requirements in customer‑friendly terms.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-01-28T08:50:27Z

+def summarize_claim(claim_text):
+    """
+    Summarize an insurance claim into key facts.
+    """
+    # Split and clean sentences
+    sentences = [s.strip() for s in claim_text.split(".") if s.strip()]
+    # Take the first two sentences
+    summary = ". ".join(sentences[:2]) + "."


The summarize_claim function has a bug: it naively splits by periods without considering periods in abbreviations, numbers, or other contexts. For example, if a claim mentions "Dr. Smith" or "123.45", these would be incorrectly split. Consider using more robust sentence tokenization methods from libraries like nltk or spaCy, or at minimum, handle common edge cases like titles (Mr., Dr., etc.) and decimal numbers.

Suggested change

def summarize_claim(claim_text):

"""

Summarize an insurance claim into key facts.

"""

# Split and clean sentences

sentences = [s.strip() for s in claim_text.split(".") if s.strip()]

# Take the first two sentences

summary = ". ".join(sentences[:2]) + "."

import re

def summarize_claim(claim_text):

"""

Summarize an insurance claim into key facts.

"""

# Protect common abbreviations and decimal numbers from being split

abbreviations = ["Mr.", "Mrs.", "Ms.", "Dr.", "Prof.", "Sr.", "Jr.", "St."]

protected_text = claim_text

for abbr in abbreviations:

placeholder = abbr.replace(".", "<ABBR_DOT>")

protected_text = protected_text.replace(abbr, placeholder)

# Protect decimal points between digits (e.g., 123.45)

protected_text = re.sub(r"(?<=\d)\.(?=\d)", "<DECIMAL_DOT>", protected_text)

# Split into sentences on sentence-ending punctuation followed by whitespace

raw_sentences = re.split(r"(?<=[.!?])\s+", protected_text.strip())

# Restore protected periods and clean sentences

sentences = []

for s in raw_sentences:

s = s.replace("<ABBR_DOT>", ".").replace("<DECIMAL_DOT>", ".").strip()

if s:

sentences.append(s)

if not sentences:

return ""

# Take the first two sentences

summary = " ".join(sentences[:2])

# Ensure the summary ends with sentence-ending punctuation

if not summary.endswith((".", "!", "?")):

summary += "."

Copilot · 2026-01-28T08:50:27Z

+def summarize_claim(claim_text):
+    """
+    Summarize an insurance claim into key facts.
+    """
+    # Split and clean sentences
+    sentences = [s.strip() for s in claim_text.split(".") if s.strip()]
+    # Take the first two sentences
+    summary = ". ".join(sentences[:2]) + "."
+    return summary
+
+# Step 2: Example claim document
+claim_document = "On December 10th, 2025, Mr. Ade filed a motor accident claim. His Toyota Corolla was rear-ended at a traffic light in Lagos. The estimated repair cost is ₦1,200,000. No injuries were reported."
+
+
+# Step 3: Run the summarizer
+print("Original Claim Document:")
+print(claim_document)
+print("\nAI-Generated Summary:")
+print(summarize_claim(claim_document))


These files are placed at the root level of the repository, which is inconsistent with the repository's structure. All other code examples in this repository are organized within numbered lesson folders (e.g., 06-text-generation-apps/python/, 09-building-image-applications/python/). Consider organizing these files within an appropriate lesson folder or creating a new lesson folder if this represents a new lesson. For reference, see the organization pattern in folders like 06-text-generation-apps/ and 09-building-image-applications/.

Copilot · 2026-01-28T08:50:27Z

+def summarize_claim(claim_text):
+    """
+    Summarize an insurance claim into key facts.
+    """
+    # Split and clean sentences
+    sentences = [s.strip() for s in claim_text.split(".") if s.strip()]
+    # Take the first two sentences
+    summary = ". ".join(sentences[:2]) + "."
+    return summary


The summarize_claim function doesn't actually use any generative AI or LLM functionality - it's just basic string manipulation that splits text by periods and takes the first two sentences. This is misleading in a repository focused on teaching Generative AI applications. Consider implementing this using an actual AI service (like OpenAI or Azure OpenAI) to demonstrate how AI can intelligently summarize claims, extract key information, or generate structured summaries. See examples in 06-text-generation-apps/python/oai-app.py for the pattern used in this repository.

Copilot · 2026-01-28T08:50:28Z

@@ -0,0 +1,24 @@
+# Generative AI in Nigerian Insurance
+
+Generative AI can help insurance companies improve efficiency, reduce fraud, and enhance customer service.  


This line has trailing whitespace (two spaces at the end). Remove the trailing spaces for cleaner code formatting.

Suggested change

Generative AI can help insurance companies improve efficiency, reduce fraud, and enhance customer service.

Generative AI can help insurance companies improve efficiency, reduce fraud, and enhance customer service.

oludarebakare · 2026-02-16T13:28:38Z

Sounds good Thanks

…

On Sat, Dec 13, 2025 at 1:03 AM github-actions[bot] < ***@***.***> wrote: *github-actions[bot]* left a comment (microsoft/generative-ai-for-beginners#1056) <#1056 (comment)> 👋 Thanks for contributing @oludarebakare <https://github.com/oludarebakare>! We will review the pull request and get back to you soon. — Reply to this email directly, view it on GitHub <#1056 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BY3LL2F26XFTAYDAPSN6KP34BNJTZAVCNFSM6AAAAACO47AZOWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTMNBYGU2TKOJWGQ> . You are receiving this because you were mentioned.Message ID: ***@***.***>

Added insurance AI use case examples

63a4a8a

github-actions Bot added the needs-review label Dec 13, 2025

github-actions Bot assigned koreyspace Dec 13, 2025

github-actions Bot requested a review from koreyspace December 13, 2025 00:02

Add Python demo for summarizing insurance claims

1c4a94e

leestott requested a review from Copilot January 28, 2026 08:45

Copilot started reviewing on behalf of leestott January 28, 2026 08:46 View session

Copilot AI reviewed Jan 28, 2026

View reviewed changes

koreyspace closed this Feb 16, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add insurance AI use case examples#1056

Add insurance AI use case examples#1056
oludarebakare wants to merge 2 commits into
microsoft:mainfrom
oludarebakare:my-first-lesson

oludarebakare commented Dec 13, 2025

Uh oh!

github-actions Bot commented Dec 13, 2025

Uh oh!

oludarebakare commented Dec 13, 2025

Uh oh!

koreyspace commented Jan 12, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Jan 28, 2026

Uh oh!

Copilot AI Jan 28, 2026

Uh oh!

Copilot AI Jan 28, 2026

Uh oh!

Copilot AI Jan 28, 2026

Uh oh!

oludarebakare commented Feb 16, 2026 via email

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

-def summarize_claim(claim_text):
-    """
-    Summarize an insurance claim into key facts.
-    """
-    # Split and clean sentences
-    sentences = [s.strip() for s in claim_text.split(".") if s.strip()]
-    # Take the first two sentences
-    summary = ". ".join(sentences[:2]) + "."
+import re
+def summarize_claim(claim_text):
+    """
+    Summarize an insurance claim into key facts.
+    """
+    # Protect common abbreviations and decimal numbers from being split
+    abbreviations = ["Mr.", "Mrs.", "Ms.", "Dr.", "Prof.", "Sr.", "Jr.", "St."]
+    protected_text = claim_text
+    for abbr in abbreviations:
+        placeholder = abbr.replace(".", "<ABBR_DOT>")
+        protected_text = protected_text.replace(abbr, placeholder)
+    # Protect decimal points between digits (e.g., 123.45)
+    protected_text = re.sub(r"(?<=\d)\.(?=\d)", "<DECIMAL_DOT>", protected_text)
+    # Split into sentences on sentence-ending punctuation followed by whitespace
+    raw_sentences = re.split(r"(?<=[.!?])\s+", protected_text.strip())
+    # Restore protected periods and clean sentences
+    sentences = []
+    for s in raw_sentences:
+        s = s.replace("<ABBR_DOT>", ".").replace("<DECIMAL_DOT>", ".").strip()
+        if s:
+            sentences.append(s)
+    if not sentences:
+        return ""
+    # Take the first two sentences
+    summary = " ".join(sentences[:2])
+    # Ensure the summary ends with sentence-ending punctuation
+    if not summary.endswith((".", "!", "?")):
+        summary += "."

		@@ -0,0 +1,24 @@
		# Generative AI in Nigerian Insurance

		Generative AI can help insurance companies improve efficiency, reduce fraud, and enhance customer service.

Conversation

oludarebakare commented Dec 13, 2025

Uh oh!

github-actions Bot commented Dec 13, 2025

Uh oh!

oludarebakare commented Dec 13, 2025

Uh oh!

koreyspace commented Jan 12, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Jan 28, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 28, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 28, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 28, 2026

Choose a reason for hiding this comment

Uh oh!

oludarebakare commented Feb 16, 2026 via email

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants