Database administrators and architects

Back up and restore data to prevent data loss

Backup and restore operations are highly procedural and rule-based. AI systems can schedule backups, verify integrity, automate restore testing, and execute recovery procedures with minimal human intervention. This task is already heavily automated in modern database platforms.

BLS evidence: The duties explicitly include backing up and restoring data to prevent data loss.

Maintain databases and update permissions

Updating permissions and routine maintenance tasks are highly structured and rule-based. AI can parse access requests, apply role-based policies, audit permission changes, and execute maintenance scripts with minimal human involvement beyond approval workflows for sensitive access changes.

BLS evidence: The duties list includes maintaining databases and updating permissions as ongoing administrative tasks.

Install upgrades and patches to fix program bugs

Installing patches and upgrades is largely procedural with well-defined steps. AI can read release notes, assess compatibility, schedule maintenance windows, and execute installations. Human review is needed for critical systems and rollback decisions, but the task is highly automatable.

BLS evidence: System DBAs are responsible for installing upgrades and patches to fix program bugs.

Ensure databases operate efficiently and without error

AI-powered monitoring tools can detect anomalies, optimize queries, tune indexes, and auto-remediate common issues with high accuracy. Human oversight remains necessary for novel failures and capacity planning decisions, but the bulk of routine operational work is increasingly automated.

BLS evidence: Database administrators ensure that systems perform as they should by monitoring database operation and providing support.

Write or debug programs for database applications

Current AI code generation tools can write stored procedures, triggers, and database application code from specifications, and debug common errors effectively. Complex business logic and performance optimization still benefit from human expertise, but AI can handle substantial portions autonomously.

BLS evidence: Application DBAs may write or debug programs and must be able to manage the applications that work with the database.

Make and test modifications to database structure when needed

AI can generate ALTER scripts, predict migration impacts, and automate testing of schema changes. However, coordinating modifications across dependent systems, timing deployments, and validating business logic preservation require human oversight, though AI handles much of the technical execution.

BLS evidence: Database administrators and architects make and test modifications to database structure when needed.

Design and build new databases for systems and applications

AI can generate database schemas, suggest normalization strategies, and produce initial designs from requirements, but complex enterprise systems require human judgment on performance trade-offs, business logic integration, and stakeholder alignment that AI cannot fully navigate autonomously.

BLS evidence: Database architects design and build new databases for systems and applications, creating models and coding new data architecture.

Plan and implement security measures to protect organizational data

AI can recommend security configurations, detect vulnerabilities, and draft access control policies, but implementing security requires understanding organizational risk tolerance, compliance nuances, and making judgment calls on security-usability trade-offs that demand human decision-making.

BLS evidence: Database administrators often are responsible for planning security measures to protect personal, proprietary, or financial information.

Identify user needs to create and administer databases

AI can analyze usage patterns and suggest database structures, but identifying true user needs requires interviewing stakeholders, understanding unstated requirements, navigating organizational politics, and translating vague business goals into technical specifications—skills where AI assists but humans lead.

BLS evidence: Database administrators and architects identify user needs to create and administer databases, ensuring data analysts and other users can easily find information.