This page is still being polished. If you have thoughts, please share them via the feedback form.
Data on this page is preliminary and may change. Please do not share or cite these figures publicly.
Foundational safety research, theoretical understanding, and scientific inquiry informing AI development.
Also in Engineering & Development
Work on AI ethics includes developing methods for integrating ethical considerations into training, evaluation, and decision-making processes, as well as techniques for mitigating harmful outputs and ensuring cultural and long-term ethical consistency.
Reasoning
Developing ethical methods and design principles for integrating considerations into training, evaluation, and decision-making processes during system development.
Ethics-aware training and fine-tuning
Research on learning from imperfect ethical datasets, applying ethics-aware data curation methods, and incorporating collective ethical principles into model design.
1.1 ModelEthical decision-making frameworks
Developing formal risk-aware, algorithmic harms assessment, and domain-specific ethical decision-making frameworks tailored for large language models and related AI systems.
2.2.1 Risk AssessmentMitigating harmful outputs
Approaches include refining models to reduce the production of dangerous, misleading, or otherwise harmful outputs, employing filtering, red-teaming, and reinforcement learning from human feedback.
1.1.2 Learning ObjectivesCultural sensitivity and contextual awareness
Techniques aim to adapt models to diverse cultural contexts and subtle social norms, ensuring that outputs remain appropriate, respectful, and aligned with local values.
1.1 ModelLong-term ethical consistency
Research explores methods for maintaining stable, ethically coherent model behavior over extended periods, including approaches to prevent drift and to preserve core ethical principles despite shifting inputs.
2.4.1 Research & FoundationsTheoretical foundations and provable safety in AI systems
Advancing the theoretical foundations of AI safety by building models and frameworks that ensure provably correct and robust behavior. These efforts span from verifiable architectures and formal verification methods to embedded agency, decision theory, incentive structures aligned with causal reasoning, and control theory.
2.4.1 Research & FoundationsTheoretical foundations and provable safety in AI systems > Building verifiable and robust AI architectures
Constructing AI systems with architectures that support formal verification and robustness guarantees, such as world models that enable safe and reliable planning, or guaranteed safe AI with Bayesian oracles. This area emphasizes simplicity and transparency to aid in provability.
1.1.4 Model ArchitectureTheoretical foundations and provable safety in AI systems > Formal verification of AI systems
Applying formal methods to verify that AI models and algorithms meet stringent safety, robustness, and performance criteria. This includes proving resilience against adversarial inputs and perturbations, and certifying conformance to specified safety properties under varying conditions.
2.2.2 Testing & EvaluationTheoretical foundations and provable safety in AI systems > Decision theory and rational agency
Establishing formal decision-making frameworks that ensure rational and safe choices by AI agents, potentially drawing on concepts like causal and evidential decision theory.
2.4.1 Research & FoundationsTheoretical foundations and provable safety in AI systems > Embedded agency
Explores how agents can model and reason about themselves and their environment as interconnected parts of a single system, addressing challenges like self-reference, resource constraints, and the stability of reasoning processes. This includes tackling problems arising from the lack of a clear boundary between the agent and its environment.
2.4.1 Research & FoundationsTheoretical foundations and provable safety in AI systems > Causal incentives
Developing frameworks that formalize how to align agent incentives with safe and desired outcomes by ensuring their causal understanding matches intended objectives. This research provides a formal language for guaranteeing safety, addressing challenges like goal misspecification, and complementing broader efforts in agent foundations and robust system design.
2.4.1 Research & FoundationsExpert Survey: AI Reliability & Security Research Priorities
O'Brien, Joe; Dolan, Jeremy; Kim, Jay; Dykhuizen, Jonah; Sania, Jeba; Becker, Sebastian; Kraprayoon, Jam; Labrador, Cara (2025)
Our survey of 53 specialists across 105 AI reliability and security research areas identifies the most promising research prospects to guide strategic AI R&D investment. As companies are seeking to develop AI systems with broadly human-level capabilities, research on reliability and security is urgently needed to ensure AI's benefits can be safely and broadly realized and prevent severe harms. This study is the first to quantify expert priorities across a comprehensive taxonomy of AI safety and security research directions and to produce a data-driven ranking of their potential impact. These rankings may support evidence-based decisions about how to effectively deploy resources toward AI reliability and security research.
Other (multiple stages)
Applies across multiple lifecycle stages
Developer
Entity that creates, trains, or modifies the AI system
Other
Risk management function not captured by the standard AIRM categories
Other
