Prompt Injection, 2022 vs Today: A Retrospective
Published:
Date: June 20, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Posts by Tags
Adversarial Fine-Tuning
Code Assistants
Does AI Make You Write Insecure Code? A User Study
Published:
Date: June 15, 2026 | Estimated Reading Time: 9 min | Author: Gustavo Sandoval
Deep Learning
Adversarial Fine-Tuning against Prompt Injection Attacks
Published:
Date: January 1, 2023 | Estimated Reading Time: 7 min | Authors: Gustavo Sandoval, Denys Fenchenko, Junyao Chen (NYU)
Evaluation
Red-Teaming Language Models, Part 3: What Changes When You Ship Open Weights
Published:
Date: August 5, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 2: What Holds on Defense
Published:
Date: August 3, 2026 | Estimated Reading Time: 5 min | Author: Gustavo Sandoval
What Counts as an Adequate Threat Model?
Published:
Date: July 31, 2026 | Estimated Reading Time: 24 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 1: The Process and the Measurement Problem
Published:
Date: July 29, 2026 | Estimated Reading Time: 15 min | Author: Gustavo Sandoval
Governance
What Counts as an Adequate Threat Model?
Published:
Date: July 31, 2026 | Estimated Reading Time: 24 min | Author: Gustavo Sandoval
Jailbreak
Red-Teaming Language Models, Part 3: What Changes When You Ship Open Weights
Published:
Date: August 5, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 2: What Holds on Defense
Published:
Date: August 3, 2026 | Estimated Reading Time: 5 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 1: The Process and the Measurement Problem
Published:
Date: July 29, 2026 | Estimated Reading Time: 15 min | Author: Gustavo Sandoval
LLM
Red-Teaming Language Models, Part 3: What Changes When You Ship Open Weights
Published:
Date: August 5, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 2: What Holds on Defense
Published:
Date: August 3, 2026 | Estimated Reading Time: 5 min | Author: Gustavo Sandoval
What Counts as an Adequate Threat Model?
Published:
Date: July 31, 2026 | Estimated Reading Time: 24 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 1: The Process and the Measurement Problem
Published:
Date: July 29, 2026 | Estimated Reading Time: 15 min | Author: Gustavo Sandoval
Mechanistic Interpretability as a Security Tool
Published:
Date: June 22, 2026 | Estimated Reading Time: 12 min | Author: Gustavo Sandoval
AI Literacy Audit and Rollout Plan
Published:
Date: June 22, 2026 | Estimated Reading Time: 8 min | Author: Gustavo Sandoval
Prompt Injection, 2022 vs Today: A Retrospective
Published:
Date: June 20, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
The Format-Reliability Gap: Diagnosing and Repairing Insecure Code Generation
Published:
Date: June 18, 2026 | Estimated Reading Time: 10 min | Author: Gustavo Sandoval
Does AI Make You Write Insecure Code? A User Study
Published:
Date: June 15, 2026 | Estimated Reading Time: 9 min | Author: Gustavo Sandoval
Adversarial Fine-Tuning against Prompt Injection Attacks
Published:
Date: January 1, 2023 | Estimated Reading Time: 7 min | Authors: Gustavo Sandoval, Denys Fenchenko, Junyao Chen (NYU)
Mechanistic Interpretability
Mechanistic Interpretability as a Security Tool
Published:
Date: June 22, 2026 | Estimated Reading Time: 12 min | Author: Gustavo Sandoval
The Format-Reliability Gap: Diagnosing and Repairing Insecure Code Generation
Published:
Date: June 18, 2026 | Estimated Reading Time: 10 min | Author: Gustavo Sandoval
Open Weights
Red-Teaming Language Models, Part 3: What Changes When You Ship Open Weights
Published:
Date: August 5, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 2: What Holds on Defense
Published:
Date: August 3, 2026 | Estimated Reading Time: 5 min | Author: Gustavo Sandoval
What Counts as an Adequate Threat Model?
Published:
Date: July 31, 2026 | Estimated Reading Time: 24 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 1: The Process and the Measurement Problem
Published:
Date: July 29, 2026 | Estimated Reading Time: 15 min | Author: Gustavo Sandoval
Prompt Injection
Prompt Injection, 2022 vs Today: A Retrospective
Published:
Date: June 20, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red Teaming
Red-Teaming Language Models, Part 3: What Changes When You Ship Open Weights
Published:
Date: August 5, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 2: What Holds on Defense
Published:
Date: August 3, 2026 | Estimated Reading Time: 5 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 1: The Process and the Measurement Problem
Published:
Date: July 29, 2026 | Estimated Reading Time: 15 min | Author: Gustavo Sandoval
Secure Code Generation
The Format-Reliability Gap: Diagnosing and Repairing Insecure Code Generation
Published:
Date: June 18, 2026 | Estimated Reading Time: 10 min | Author: Gustavo Sandoval
Security
Red-Teaming Language Models, Part 3: What Changes When You Ship Open Weights
Published:
Date: August 5, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 2: What Holds on Defense
Published:
Date: August 3, 2026 | Estimated Reading Time: 5 min | Author: Gustavo Sandoval
What Counts as an Adequate Threat Model?
Published:
Date: July 31, 2026 | Estimated Reading Time: 24 min | Author: Gustavo Sandoval
Red-Teaming Language Models, Part 1: The Process and the Measurement Problem
Published:
Date: July 29, 2026 | Estimated Reading Time: 15 min | Author: Gustavo Sandoval
Mechanistic Interpretability as a Security Tool
Published:
Date: June 22, 2026 | Estimated Reading Time: 12 min | Author: Gustavo Sandoval
AI Literacy Audit and Rollout Plan
Published:
Date: June 22, 2026 | Estimated Reading Time: 8 min | Author: Gustavo Sandoval
Prompt Injection, 2022 vs Today: A Retrospective
Published:
Date: June 20, 2026 | Estimated Reading Time: 11 min | Author: Gustavo Sandoval
The Format-Reliability Gap: Diagnosing and Repairing Insecure Code Generation
Published:
Date: June 18, 2026 | Estimated Reading Time: 10 min | Author: Gustavo Sandoval
Does AI Make You Write Insecure Code? A User Study
Published:
Date: June 15, 2026 | Estimated Reading Time: 9 min | Author: Gustavo Sandoval
Adversarial Fine-Tuning against Prompt Injection Attacks
Published:
Date: January 1, 2023 | Estimated Reading Time: 7 min | Authors: Gustavo Sandoval, Denys Fenchenko, Junyao Chen (NYU)
Steering
Mechanistic Interpretability as a Security Tool
Published:
Date: June 22, 2026 | Estimated Reading Time: 12 min | Author: Gustavo Sandoval
Threat Modeling
What Counts as an Adequate Threat Model?
Published:
Date: July 31, 2026 | Estimated Reading Time: 24 min | Author: Gustavo Sandoval
User Study
Does AI Make You Write Insecure Code? A User Study
Published:
Date: June 15, 2026 | Estimated Reading Time: 9 min | Author: Gustavo Sandoval
Vibe Coding
AI Literacy Audit and Rollout Plan
Published:
Date: June 22, 2026 | Estimated Reading Time: 8 min | Author: Gustavo Sandoval