Research Scientist, AGI Safety and Alignment, DeepMind
Research Scientist, AGI Safety and Alignment, DeepMind Overview
| Company Name | |
| Job Role | Research Scientist, AGI Safety and Alignment, DeepMind |
| Qualifications | Not Specified |
| Category | IT Jobs |
| Job Type | Full Time |
| Location | London |
DeepMind, a renowned AI research organization under Google, is seeking a dedicated Research Scientist to focus on the safety and alignment of artificial general intelligence (AGI). This role involves pioneering research to reduce the risks associated with AGI and future superintelligent systems, ensuring that these powerful AI models are aligned with human values and safety standards.
Key Responsibilities
- Innovate and develop new methods for aligning AGI, addressing potential failure modes, and applying scalable alignment techniques to frontier models.
- Design and implement control systems for AGI, ensuring these controls are effective in real-world deployment.
- Conduct interpretability research to understand the internal workings of AI systems, promoting transparency and safety.
- Collaborate with product teams to translate research into practical applications and ensure safety measures are integrated into AI products.
- Investigate and develop new alignment techniques, including evaluation methods and tools to detect misalignment.
- Explore training methodologies such as debate to align superhuman AI, and develop tools to monitor and measure model alignment.
- Assess potential risks posed by AI models through threat modeling and safety evaluations, providing insights to leadership.
- Create safety tools and AI assistance to accelerate research and understanding of AI safety issues.
- Advise executive teams on AI safety risks, based on comprehensive threat models and safety evaluations.
Qualifications
- A bachelorĂ¢??s degree in Computer Science, Software Engineering, or a related field, or equivalent practical experience.
- Minimum of three years of experience in software development, machine learning engineering, or research related to ML.
- Experience working with research teams on AI safety or alignment projects.
- Preferred: Experience in applied research to improve AI safety and alignment, especially with large models.
- Preferred: Hands-on experience with training large models, including supervised fine-tuning and reinforcement learning from human feedback (RLHF).
About the Role
The team focuses on reducing existential risks from AGI and future superintelligence by researching novel alignment techniques, safety evaluation, interpretability, and control mechanisms. The role involves working on cutting-edge AI safety challenges, including model forensics, interpretability, and developing tools for safety assessment. The team also advises leadership on safety risks and helps develop frameworks for threat modeling and safety evaluation.
DeepMind emphasizes the importance of safety, ethics, and responsible AI development, aiming to benefit society through scientific discovery and technological advancement. The organization values diversity and is committed to equal opportunity employment.
Additional Information
Applicants are encouraged to complete accommodation requests if needed. English proficiency is required for global collaboration. Google does not accept agency resumes and is not responsible for fees related to unsolicited applications. Equity grants are discretionary and subject to approval by Alphabet Inc., with no guaranteed rights or impact on statutory payments.
Google is dedicated to fostering an inclusive environment and adheres to policies promoting equal opportunity and non-discrimination across all roles.
Degree Requirement: Not Specified
Visa Sponsorship May be
To apply for this job please visit www.google.com.