SecAI+ Secure AI Development and MLSecOps

Question 1

What is MLSecOps primarily focused on?

Accepted Answer

Integrating security practices into the ML development lifecycle

Answer

Using ML exclusively to detect network intrusions

Answer

Replacing traditional DevOps pipelines with AI automation

Answer

Automating vulnerability scanning using neural networks only

Question 2

Which phase of the ML pipeline is most vulnerable to data poisoning attacks?

Accepted Answer

Training data collection and preparation

Answer

Model inference and serving

Answer

Model deployment to production

Answer

Model evaluation and benchmarking

Question 3

What is the primary purpose of model versioning in a secure ML pipeline?

Accepted Answer

To track changes and enable rollback if a model is compromised

Answer

To accelerate model training iterations

Answer

To share models publicly for community transparency

Answer

To reduce inference computational costs over time

Question 4

Which practice best ensures the integrity of ML artifacts during a CI/CD build pipeline?

Accepted Answer

Cryptographic signing and hash verification of model files

Answer

Randomizing training data splits before each build

Answer

Using larger training datasets for every release

Answer

Disabling model logging during training to reduce overhead

Question 5

What does a 'model card' document primarily provide in secure AI development?

Accepted Answer

Transparency about a model's capabilities, limitations, and intended use cases

Answer

A legal software license governing AI system usage rights

Answer

A performance benchmark comparing the model against competitors

Answer

An API specification for integrating the model into applications

Question 6

In a secure ML pipeline, what is the primary security role of a feature store?

Accepted Answer

A centralized repository for managing, versioning, and sharing ML features consistently

Answer

A database storing only raw, unprocessed data for compliance

Answer

A tool for visualizing model predictions in real time

Answer

A component that encrypts model weights during inference requests

Question 7

Which security control is most critical when exposing ML models as REST API endpoints?

Accepted Answer

Authentication, authorization, and rate limiting on model endpoints

Answer

Using the largest available model size for accuracy

Answer

Making model weights publicly available for third-party audits

Answer

Disabling request logging to protect user privacy from the start