A person at a computer. A little digital robot is hovering near his fingers.

Georgia Tech Researchers Find a Way to Reduce AI Chatbots’ Reinforcement of Delusions

Georgia Tech researchers found that instruction changes can reduce chatbots’ reinforcement of delusional beliefs.

A group photo of students and faculty standing around lab equipment found on a ship.

When a Ship’s Systems Lie, Cybersecurity Gets Complicated

Maritime cybersecurity training leaves sailors vulnerable to attacks that manipulate ship systems without obvious failures. The Georgia Tech study identifies weaknesses in preparation and threat recognition among military mariners.

A red robot in front of a yellow background emits a speech bubble that indicates harmful output

Can We Detect Harmful AI Before It’s Too Late? Researchers Search for Warning Signs

A Georgia Tech Ph.D. student is investigating whether internal signals in large language models could help identify harmful behavior before it reaches users.

A man points to a rack of CPUs. He is talking to a student who is slighlty out of focus.

Deleted AI Chats May Leave Clues Behind

Researchers find that deleted local AI chats may remain recoverable from a phone’s memory through a new tool called ORISA.

Georgia Tech Receives $5 Million From Google.org to Expand Access to AI for Scientific Discovery

Georgia Tech Receives $5 Million From Google.org to Expand Access to AI for Scientific Discovery

The three-year initiative gives researchers greater access to AI tools, training, and computing power to help accelerate scientific breakthroughs.

Mark Riedl

New Neural Framework Could Make LLMs More Empathetic

Georgia Tech researchers have developed a new neural framework that examines how different training sources influence empathy and social reasoning within large language models.

Photo of Buzz the mascot leaning down toward a small humanoid robot wearing a Georgia Tech t-shirt

Georgia Tech and Amazon Launch Joint Science Hub 

The collaboration will support research in robotics, industrial and systems engineering, and aerospace technology while fostering innovation in foundational and emerging technologies. 

In the movie ‘War Games’ an AI plays a game with the real threat of triggering nuclear Armageddon. Courtesy of United Artists/Amazon MGM Studios

The ‘WarGames’ Problem: Computer Science Has Long Understood What it Takes to Keep AI Under Control

AI agents don’t go rogue. That’s something only humans do.

Saman Zonouz is a Georgia Tech associate professor and lead researcher for the DerGuard project.

GT Cybersecurity Expert Helping to Chart the Future of AI on the Energy Grid

Georgia Tech researcher Saman Zonouz is helping shape resilient AI practices for the electricity grid. His work connects workshop planning with research on AI-enabled grid protection and emerging vulnerabilities.

Zsolt Kira sitting in office chair surrounded by old computers

New Benchmark Measures Memory Capabilities of VLM-Powered Robots

To improve robot performance and advance memory capabilities needed for reliable home use, Georgia Tech researchers have developed a new benchmark to evaluate the memory capabilities of vision-language models (VLMs).