Tag: AI code vulnerabilities

Databricks AI Red Team Findings: How AI-Generated Game and Parser Code Can Be Exploited

Databricks AI red team uncovered critical vulnerabilities in AI-generated game and parser code, showing how prompt injection, data leakage, and hallucinations can be exploited. These aren't theoretical risks-they're happening in real systems today.

Model Distillation for Generative AI: Smaller Models with Big Capabilities

Dec, 3 2025
Optimizing Attention Patterns for Domain-Specific Large Language Models

Oct, 10 2025
Vision-First vs Text-First Pretraining: Which Path Leads to Better Multimodal LLMs?

Nov, 27 2025
Batched Generation in LLM Serving: How Request Scheduling Shapes Output Speed and Quality

Oct, 12 2025
Code Execution as a Tool for Large Language Model Agents: How AI Systems Run Code to Solve Real Problems

Oct, 15 2025

Tag: AI code vulnerabilities

Databricks AI Red Team Findings: How AI-Generated Game and Parser Code Can Be Exploited

Recent Post

Model Distillation for Generative AI: Smaller Models with Big Capabilities

Optimizing Attention Patterns for Domain-Specific Large Language Models

Vision-First vs Text-First Pretraining: Which Path Leads to Better Multimodal LLMs?

Batched Generation in LLM Serving: How Request Scheduling Shapes Output Speed and Quality

Code Execution as a Tool for Large Language Model Agents: How AI Systems Run Code to Solve Real Problems

Categories

Archives