Pradip
03 May, 2025
0 Comments
6 Mins Read

What’s New in Oracle 23ai? What are its Latest Features?

My Journey from Oracle 19c to 23ai….still continues!

As someone who has spent years teaching and training professionals on Oracle 19c, I can confidently say that staying updated with Oracle’s latest releases is both exciting and challenging. Every time a new version drops, I find myself in a love-hate relationship with it, excited to explore the new features but also slightly panicked about rewriting all my training materials and upgrading my knowledge!

With Oracle 23ai now being the Long-Term Support (LTS) release, I knew I had to dive deep into it. So, I’ve been reading, experimenting, and learning more about 23ai in whatever little free time I get. But let me be real with you, upgrading myself isn’t easy when I’m constantly juggling continuous training batches, handling students’ doubts, and managing real-world Oracle DBA challenges.

That said, learning new things gives me a kick! There’s nothing more thrilling than discovering a game-changing feature that could save hours of DBA work or make a developer’s life easier. And trust me, Oracle 23ai is packed with such exciting features that will make you rethink how you manage databases.

So, in this article, I’m going to break down the best new features of Oracle 23ai in a way that makes sense to you. No boring documentation-style explanations, just real-world use cases, practical code snippets, and insights from my own learning experience.

Now, grab a cup of coffee (or chai, if you prefer), and let’s explore what’s new in Oracle 23ai, and why you should care to learn it!

1. JSON Relational Duality – The SQL vs. NoSQL Debate is Over!

If you’ve ever had to choose between a relational database (SQL) and a document database (NoSQL), you know how frustrating it can be. Developers love JSON for flexibility, while DBAs love relational models for data integrity and query power. Oracle 23ai solves this problem!

With JSON Relational Duality, you can now:

Store and access the same data as both JSON and relational format simultaneously.
Use SQL queries on JSON data without conversion overhead.
Eliminate the need for complex ETL processes between NoSQL and SQL databases.

Example Use Case – E-commerce Catalog

Imagine you’re working for Amazon. Your product catalog is stored in JSON for faster API responses, but your finance team needs SQL for generating revenue reports.

With Oracle 23ai, you don’t need separate storage for these two formats anymore! Your backend APIs can read/write JSON, while your reporting team runs SQL queries on the same data.

Code Example: Insert JSON and Query It with SQL

-- Creating a table with JSON Relational Duality
CREATE TABLE products (
    product_id NUMBER PRIMARY KEY,
    product_data JSON RELATIONAL DUALITY
);

-- Inserting JSON data
INSERT INTO products VALUES (1, 
    '{ "name": "iPhone 15", "price": 799, "stock": 50 }'
);

-- Querying JSON data using SQL
SELECT product_data.name, product_data.price
FROM products
WHERE product_data.stock > 10;

Mind-blowing, right? You get NoSQL speed with SQL power, without needing a separate database.

2. True Immutable Tables – Tamper-Proof Data Storage

Imagine a world where nobody can alter your data, not even DBAs. Sounds scary? Well, it’s actually super useful for industries like banking, healthcare, and legal compliance.

With Oracle 23ai Immutable Tables, you get:

Tamper-proof records (Once written, data can’t be modified or deleted.)
Regulatory compliance(Great for audit logs, financial transactions, and patient records.)
Better security (Protects against accidental or malicious data changes.)

Example Use Case – Banking Transactions

A bank like HDFC processes millions of transactions daily. If someone modifies past transactions, it could lead to fraud.

With Immutable Tables, transaction records are 100% tamper-proof, ensuring complete data integrity.

Code Example: Creating an Immutable Table

-- Creating an immutable table
CREATE IMMUTABLE TABLE bank_transactions (
    txn_id NUMBER PRIMARY KEY,
    account_no NUMBER,
    txn_amount NUMBER,
    txn_date DATE
) NO DROP NO DELETE;

-- Inserting a transaction
INSERT INTO bank_transactions VALUES (101, 1234567890, 5000, SYSDATE);

-- Trying to update or delete (This will fail!)
UPDATE bank_transactions SET txn_amount = 10000 WHERE txn_id = 101;
DELETE FROM bank_transactions WHERE txn_id = 101;

Immutable Tables ensure that once data is inserted, it stays forever!

3. Operational Property Graphs – Native Graph Queries Inside Oracle

Graph databases have been gaining popularity in recent years. Facebook, LinkedIn, and Netflix all use graph databases to model social connections, recommendations, and fraud detection.

Oracle 23ai brings graph capabilities nativelyinto the database! You can now:

Store and analyze relationships without needing Neo4j or other graph databases.
Use SQL-like syntax to perform graph queries.
Improve fraud detection, supply chain analysis, and recommendation engines.

Example Use Case – Fraud Detection in Banking

Banks like ICICI use graph models to detect fraud patterns. If a credit card is used in Delhi at 10 AM and then in Mumbai at 10:15 AM, something shady is going on!

Code Example: Finding Friends of Friends in a Social Network

-- Creating a graph table
CREATE PROPERTY GRAPH social_network 
    VERTEX TABLE users
    EDGE TABLE friendships
    SOURCE KEY (user_id)
    DESTINATION KEY (friend_id);

-- Finding friends of friends (2nd-degree connections)
SELECT friend_of_friend 
FROM TABLE(SEARCH_GRAPH(
    social_network 
    START WITH (user_id = 101) 
    CONNECT BY NOCYCLE NODES(2)
));

Graph queries can detect fraud, analyze supply chains, and build recommendation engines!

4. AI-Driven Automatic Index Optimization

How many times have you seen a DBA spending hours optimizing indexes, only to see the query performance degrade?

Well, Oracle 23ai now uses AI to optimize indexes automatically!

It learns query patternsand creates/drops indexes dynamically.
No more over-indexing(which slows down updates).
No more under-indexing (which slows down queries).

Example: Let Oracle Handle Indexing for You!

-- Enabling automatic indexing
ALTER SESSION SET AUTO_INDEX_MODE = ON;

-- Checking auto-created indexes
SELECT * FROM DBA_AUTO_INDEXES;

DBAs, you can finally take that vacation without worrying about indexes!

5. Passwordless Authentication – Say Goodbye to Password Hassles!

We’ve all been there, forgot your password?

Oracle 23ai now allows passwordless authentication, so you can log in using:

Biometric authentication(Face ID, fingerprint)
Hardware security keys(YubiKey, smart cards)
OAuth-based logins

This reduces the risk of password leaks and improves security compliance for enterprises.

Example: Setting Up Passwordless Authentication

-- Creating a user with passwordless login
CREATE USER admin IDENTIFIED EXTERNALLY AS 'biometric_auth';
GRANT CONNECT, DBA TO admin;

No more forgotten passwords = fewer helpdesk tickets!

Why Should You Upgrade to Oracle 23ai?

Still on Oracle 19c or 21c? Let me make this easy for you:

Want faster query performance?→ Upgrade!
Need better security & compliance?→ Upgrade!
Want developer-friendly features?→ Upgrade!
Like doing things the hard way? → Stay where you are… but why suffer?

Conclusion

Oracle 23ai introduces a significant evolution in database management. Whether it’s JSON Relational Duality, Immutable Tables, Graph Queries, AI-Driven Indexing, or Passwordless Authentication, this version brings features that will make your work more efficient, secure, and streamlined. If you’re still on Oracle 19c or 21c, it’s a great time to think about upgrading and exploring the new possibilities that Oracle 23ai offers.

But I understand, adapting to new technologies can be a challenge. That’s why at Learnomate Technologies, we offer top-notch, hands-on training for Oracle 23ai, ensuring that you stay up to date and make the most of these innovations. Our courses are designed to break down complex topics into easy-to-understand and practical lessons.

If you want to see real-time demos, tutorials, and expert insights, check out our YouTube channel: www.youtube.com/@learnomate

For detailed course information, training schedules, and career guidance, visit our website: www.learnomate.org

Stay connected with me on LinkedIn for more updates on Oracle databases, IT careers, and industry trends: Follow me on LinkedIn

If you’re looking to dive deeper into different technologies, feel free to explore our blog page: Read our blogs

Let’s keep learning and growing together! If you have any questions about Oracle 23ai, feel free to reach out.

Happy Reading!

ANKUSH😎