Red teaming and adversarial testing in Prompt Engineering / GenAI - Model Pipeline Trace

Model Pipeline - Red teaming and adversarial testing

This pipeline shows how red teaming and adversarial testing help find weaknesses in AI models by feeding tricky inputs and checking model responses.

Data Flow - 6 Stages

1Data in

1000 rows x 10 columns→Collect normal and adversarial examples (inputs designed to fool the model)→1000 rows x 10 columns

Normal input: 'The cat sat on the mat.' Adversarial input: 'The c@t s@t on the m@t.'

↓

2Preprocessing

1000 rows x 10 columns→Clean text, tokenize, and convert to numbers→1000 rows x 50 tokens

Input text converted to token IDs like [12, 45, 78, ...]

↓

3Feature Engineering

1000 rows x 50 tokens→Embed tokens into vectors→1000 rows x 50 tokens x 128 features

Token 'cat' becomes a 128-dimensional vector

↓

4Model Trains

800 rows x 50 tokens x 128 features→Train model on normal and adversarial data→Trained model

Model learns to classify inputs correctly despite adversarial noise

↓

5Metrics Improve

Validation set 200 rows x 50 tokens x 128 features→Evaluate accuracy and robustness→Accuracy: 85%, Robustness score: 78%

Model correctly classifies 85% of inputs including adversarial ones

↓

6Prediction

1 row x 50 tokens x 128 features→Model predicts label for new input→Prediction: 'Safe' or 'Adversarial'

Input: 'The c@t s@t on the m@t.' Output: 'Adversarial'

Training Trace - Epoch by Epoch

Loss
1.2 |*       
0.9 | **     
0.7 |  ***   
0.5 |    ****
0.4 |     *****
     ----------------
      1  2  3  4  5  Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	1.2	0.55	Model starts learning but struggles with adversarial examples
2	0.9	0.65	Loss decreases, accuracy improves as model adapts
3	0.7	0.75	Better handling of adversarial inputs
4	0.5	0.82	Model robustness improves
5	0.4	0.85	Training converges with good accuracy and robustness

Prediction Trace - 5 Layers

Layer 1: Input preprocessing

Layer 2: Embedding layer

Layer 3: Neural network layers

Layer 4: Output layer

Layer 5: Prediction decision

Model Quiz - 3 Questions

What is the main goal of adversarial testing in this pipeline?

A To speed up model training

B To increase the size of the training data

C To find inputs that trick the model

D To reduce the number of model layers

Key Insight

Red teaming and adversarial testing help models learn to recognize tricky inputs, improving their safety and reliability by exposing weaknesses during training.

Practice

1. What is the main goal of red teaming in AI?

easy

A. To find weaknesses by testing with tricky inputs

B. To train the AI model with more data

C. To improve the speed of the AI model

D. To reduce the size of the AI model

Solution

1. Step 1: Understand red teaming purpose

   Red teaming is about testing AI models with challenging inputs to find weaknesses.

2. Step 2: Compare options

   Only To find weaknesses by testing with tricky inputs matches this goal; others relate to training, speed, or size, which are unrelated.

3. Final Answer:

   To find weaknesses by testing with tricky inputs -> Option A

4. Quick Check:

   Red teaming = find weaknesses [OK]

Hint: Red teaming means testing for weaknesses with tricky inputs [OK]

Common Mistakes:

• Confusing red teaming with training
• Thinking it improves speed or size
• Assuming it fixes bugs automatically

2. Which of the following is the correct way to describe an adversarial example?

easy

A. A normal input that the model handles well

B. A training example used to improve accuracy

C. A random input unrelated to the task

D. An input designed to confuse the AI model

Solution

1. Step 1: Define adversarial example

   An adversarial example is a carefully crafted input meant to confuse or trick the AI model.

2. Step 2: Match definition to options

   An input designed to confuse the AI model matches this exactly; others describe normal, random, or training inputs.

3. Final Answer:

   An input designed to confuse the AI model -> Option D

4. Quick Check:

   Adversarial example = tricky input [OK]

Hint: Adversarial examples are tricky inputs to confuse AI [OK]

Common Mistakes:

• Thinking adversarial means normal or random input
• Confusing training data with adversarial examples
• Assuming adversarial examples improve model accuracy

3. Consider this Python code snippet for adversarial testing:

def test_model(model, inputs):
    results = []
    for inp in inputs:
        pred = model.predict(inp)
        if pred == 'safe':
            results.append(True)
        else:
            results.append(False)
    return results

inputs = ['normal', 'tricky', 'normal']
class DummyModel:
    def predict(self, x):
        return 'safe' if x == 'normal' else 'unsafe'

model = DummyModel()
print(test_model(model, inputs))

What is the output?

medium

A. [False, True, False]

B. [True, True, True]

C. [True, False, True]

D. [False, False, False]

Solution

1. Step 1: Understand model predictions

   The DummyModel returns 'safe' for 'normal' inputs and 'unsafe' for others.

2. Step 2: Evaluate each input

   Inputs are ['normal', 'tricky', 'normal']. Predictions: 'safe', 'unsafe', 'safe'. Results: True, False, True.

3. Final Answer:

   [True, False, True] -> Option C

4. Quick Check:

   Predictions match results [OK]

Hint: Check each input prediction carefully [OK]

Common Mistakes:

• Mixing up 'safe' and 'unsafe' outputs
• Assuming all inputs are safe
• Ignoring the else condition

4. This code tries to detect adversarial inputs but has a bug:

def detect_adversarial(inputs, model):
    flagged = []
    for i in inputs:
        if model.predict(i) == 'safe':
            flagged.append(i)
    return flagged

class Model:
    def predict(self, x):
        return 'unsafe' if x == 'tricky' else 'safe'

inputs = ['normal', 'tricky', 'normal']
print(detect_adversarial(inputs, Model()))

What is the bug?

medium

A. The model.predict method is missing

B. It flags safe inputs instead of unsafe ones

C. The inputs list is empty

D. The function returns a boolean instead of a list

Solution

1. Step 1: Analyze detection logic

   The function flags inputs where model.predict returns 'safe'.

2. Step 2: Check model behavior

   Model returns 'unsafe' for 'tricky', 'safe' otherwise. So safe inputs are flagged, which is wrong.

3. Final Answer:

   It flags safe inputs instead of unsafe ones -> Option B

4. Quick Check:

   Flagging logic reversed [OK]

Hint: Check if flagged inputs match unsafe cases [OK]

Common Mistakes:

• Assuming model.predict is missing
• Thinking inputs list is empty
• Confusing return types

5. You want to improve an AI chatbot's safety by using red teaming and adversarial testing. Which combined approach is best?

hard

A. Use tricky inputs to find weaknesses, then retrain with those examples

B. Ignore tricky inputs and focus on normal conversation data

C. Only test with random inputs and fix errors found

D. Reduce model size to avoid complex errors

Solution

1. Step 1: Understand red teaming and adversarial testing roles

   They find weaknesses by using tricky inputs to test the model.

2. Step 2: Combine testing with retraining

   After finding weaknesses, retraining with those examples improves safety and reliability.

3. Final Answer:

   Use tricky inputs to find weaknesses, then retrain with those examples -> Option A

4. Quick Check:

   Test + retrain = better safety [OK]

Hint: Test with tricky inputs, then retrain to fix weaknesses [OK]

Common Mistakes:

• Only testing without retraining
• Ignoring tricky inputs
• Thinking smaller models fix safety