Research Fellow / PhD Candidate in AI Safety & Alignment

Overview

Research Fellow / PhD Candidate in AI Safety & Alignment

Institution : AI Security and Safety Department, Simula Cybersecurity, Oslo, Norway

Position : Fully-funded PhD Candidate (3-4 years) or Postdoctoral Fellow (3 years)

What We Are Looking For

We are seeking candidates to lead our efforts in AI safety red teaming and to define methodologies for preventing unintended, harmful behaviors in open-source AI models. The role focuses on foundational challenges of safety, including mitigating algorithmic bias, ensuring controllability, and transparency of systems.

This is an opportunity to shape the future of trustworthy AI in Oslo, Norway.

Research Focus Areas

• Red Teaming Framework for AI Safety : Develop a comprehensive, systematic framework for assessing and mitigating AI safety risks, including scenario-based evaluation for closed and open-source models.
• Bias and Misalignment Probing : Create scalable techniques to automatically detect and measure subtle biases and misalignments in large-scale AI models.
• Behavioral Analysis & Interpretability (XAI) : Investigate internal model behavior to understand root causes of unintended actions and advance explainable AI for safety.
• Ethical and Societal Impacts : Document safety flaws from AI-human interactions, including misinformation, social manipulation, or loss of control.

Responsibilities and Qualifications

What We Offer

Application Process

Application Deadline : October 7, 2025.

Inquiries : Jostein Jensen, Director of Simula Cyber Security :

We look forward to receiving your application!

Simula Research Laboratory uses background checks in recruitment. Information about applicants may be included in public applicant lists per Norwegian law. Accepted file types : pdf, doc, docx, txt, rtf.

#J-18808-Ljbffr