About us

AIDX focuses on
building trust in
AI systems.

Founded by AI researchers and technology leaders, AIDX brings deep expertise in AI testing, software reliability, security, and governance. We are currently active in the Singapore market, working with government agencies, research institutions, and industry partners to support responsible AI adoption.

As AI moves into high-impact industries, we help organizations understand, evaluate, and manage the risks behind real-world AI deployment.

Meet Our Team

A unique team of AI technology and regulatory leaders.

Dr. Yifan Jia portrait

Dr. Yifan Jia

CEO

Adjunct fellow at SUTD

SG100 Women in Tech 2025 award

Ph.D. in AI security with CPS

TECHStrength

Research experience with publications in top conferences and journals

  • ACL 2026 Rendering Data Unlearnable by Exploiting LLM Alignment Mechanisms
  • ICLR 2026 Where Did It Go Wrong? Attributing Undesirable LLM Behaviors via Representation Gradient Tracing
  • NDSS 2026 Q-MLLM: Vector Quantization for Robust Multimodal Large Language Model Security
  • NDSS 2026 Rounding-Guided Backdoor Injection in Deep Learning Model Quantization
  • AAAI 2026 Towards Provably Unlearnable Examples via Bayes Error Optimization
  • ICLR 2025 Democratic Training Against Universal Adversarial Perturbations
  • ICML 2025 CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization
  • OOPSLA 2025 Correct-By-Construction: Certified Individual Fairness through Neural Network Training
  • ICSE 2025 FixDrive: Automatically Repairing Autonomous Vehicle Driving Behaviour for $0.08 per Violation
  • EMNLP 2025 Zero-Shot Defense Against Toxic Images via Inherent Multimodal Alignment in LVLMs
  • EMNLP 2024 Defending Large Language Models Against Jailbreak Attacks via Layer-specific Editing
  • CAV 2024 Certified Robust Accuracy of Neural Networks Are Bounded due to Bayes Errors
Presentation at an academic conference

Collaborations with top research team

  • A white-box deep learning model copyright protection method based on neuron output.
  • A black-box deep learning model copyright protection method based on adversarial sample fingerprints.
  • A robust watermark forgetting verification method based on adversarial samples in black-box scenarios in federated learning.
Zhejiang University
Singapore Management University
National University of Singapore
Australian National University

Partners

Trusted by Industry Leaders and Government Bodies

{

Why Trust Us

}
Top Academic Background
9+Years
50+Publications

ACM SIGSOFT
Distinguished

Trusted by Singapore Government

IMDA AI Verify Foundation · Global AI Assurance Sandbox.

View Case Study

In Front of Regulation

ISO · IEEE · Singapore National Standards.

We don't just follow the rules — we help write them.

Events

AIDX brings more industry impact.

Signing ceremony — MOU with Synapxe

Officially signed an addendum to the MOU with Synapxe.

SG100 Women in Tech event

SG100 Women in Tech.

Presentation at ICSE 2026

Our work presented at ICSE 2026

Global ICT & CET Standards Program

Represented Singapore at the Global ICT & CET Standards Program.

AI training session for MCCY

AI training session at SUTD for the Ministry of Culture, Community and Youth (MCCY)

Singapore Computer Society event

Singapore Computer Society.

Trustworthy AI Forum panel discussion

Trustworthy AI (TrAI) Forum Panel Discussion

NTU–CKGSB ASEAN-RCEP Leadership Programme

NTU–CKGSB ASEAN-RCEP Leadership Programme