DARPA and IBM are ensuring that anyone can protect their AI systems from hackers
News
Mike Murphy
Data and AI Security
As organizations move to the hybrid cloud, they must protect sensitive data and comply with regulations that allow them to take advantage of AI. We’re designing systems to monitor and protect data, building trust in AI through robust evaluation, certification, and hardening against attacks.
Our work
An AI model trained on data that looks real but won’t leak personal information
ResearchKim MartineauAn open-source toolkit for debugging AI models of all data types
Technical noteKevin Eykholt and Taesung LeeSimplifying cloud security policies with AI
Technical noteJulian Stephen and Shriti PriyaProtect yourself from bad neighbors in the cloud
Technical noteMichael V. Le, Salman Ahmed, Dan Williams, and Hani JamjoomAI diffusion models can be tricked into generating manipulated images
NewsKim Martineau- See more of our work on Data and AI Security
Publications
Protecting Users From Themselves: Safeguarding Contextual Privacy in Interactions with Conversational Agents
- Ivoline Ngong
- Swanand Ravindra Kadhe
- et al.
- 2025
- ACL 2025
Think Again! The Effect of Test-Time Compute on Preferences, Opinions, and Beliefs of Large Language Models
- George Kour
- Itay Nakash
- et al.
- 2025
- ACL 2025
PATHE: A Privacy-Preserving Mass Spectrometry Database Pattern Search Platform with Fully Homomorphic Encryption
- Xuan Wang
- Minxuan Zhou
- et al.
- 2025
- DAC 2025
A Near-DRAM Accelerator for Compiler-Generated Fully Homomorphic Encryption Applications
- 2025
- ISCA 2025
IBM’s SARA SoC/SiP Project: Application-Driven High Level View
- Pradip Bose
- 2025
- ISCA 2025
Challenges and Remedies of Domain-Specific Classifiers as LLM Guardrails: Self-Harm as a Case Study
- Bing Zhang
- Guang-Jie Ren
- 2025
- NAACL 2025
IBM Solution: IBM Cloud Pak for Data
Our research is regularly incorporated into new security features for IBM Cloud Pak for Data.
Learn more