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.
Techniques to remove, bound, or modify learned model capabilities post-training.
Also in Model
Restricting risky capabilities of deployed models, such as advanced autonomy (e.g., self-assigning new sub-goals, executing long-horizon tasks) or tool use functionalities (e.g., function calls, web browsing).
Multiple experts agreed that capability restrictions could be effective, particularly for chemical, biological, radiological, and nuclear (CBRN) risks; and critical infrastructure disruptions. They noted that limiting advanced autonomy and tool use functionalities could significantly reduce these risks. However, several experts also expressed concerns about the feasibility of implementing such restrictions, citing economic incentives and the difficulty of isolating specific capabilities. Some experts mentioned that capability restrictions might be less effective for addressing bias and discrimination risks or negative effects on democratic processes, as these issues are not necessarily tied to advanced capabilities. Several experts pointed out that restricting capabilities could potentially drive development underground or be easily circumvented by skilled programmers. One expert emphasised the importance of maintaining human oversight and control, while another suggested that restrictions should be risk-scenario based rather than function-based. A single expert raised concerns about potential censorship and over-politicisation of AI if restrictions are too broad. Several experts noted the challenge of balancing risk mitigation with preserving beneficial uses of AI, particularly in research and economic applications. The need for international agreements and enforcement was mentioned by one expert.
Reasoning
Removes or restricts advanced autonomy and tool-use capabilities from deployed models post-training.
Pre-deployment risk assessments
Comprehensive risk assessments before deployment that would assess reasonably foreseeable misuse and include dangerous capability evaluations that incorporate post-training enhancements and collaborations with domain experts. Risk assessments would inform deployment decisions.
2.2.1 Risk AssessmentThird party pre-deployment model audits
External pre-deployment assessment to provide a judgment on the safety of a model. Auditors, which could be governments or independent third parties, would receive access to a fine-tuning API for testing, or further appropriate technical means.
2.2.3 Auditing & ComplianceExternal assessment of testing procedure
Bringing in external AI evaluation firms before deployment to assess and red-team the company's execution of dangerous capabilities evaluations.
2.2.2 Testing & EvaluationVetted researcher access
Giving good faith, public interest evaluation researchers access to black-box research APIs that provide technical and legal safe harbours to limit barriers imposed by usage policy enforcement, logging, and stringent terms of service.
2.3.1 Deployment ManagementAdvanced model access for vetted external researchers
Examples of advanced access rights could include any of the following: increased control over sampling, access to fine-tuning functionality, the ability to inspect and modify model internals, access to training data, or additional features like stable model versions.
2.2.2 Testing & EvaluationData curation
Careful data curation prior to all development stages (including fine-tuning) to filter out high-risk content and ensure the training data is sufficiently high-quality.
1.1.1 Training DataEffective Mitigations for Systemic Risks from General-Purpose AI
Uuk, Risto; Brouwer, Annemieke; Schreier, Tim; Dreksler, Noemi; Pulignano, Valeria; Bommasani, Rishi (2024)
The systemic risks posed by general-purpose AI models are a growing concern, yet the effectiveness of mitigations remains underexplored. Previous research has proposed frameworks for risk mitigation, but has left gaps in our understanding of the perceived effectiveness of measures for mitigating systemic risks. Our study addresses this gap by evaluating how experts perceive different mitigations that aim to reduce the systemic risks of general-purpose AI models. We surveyed 76 experts whose expertise spans AI safety; critical infrastructure; democratic processes; chemical, biological, radiological, and nuclear risks (CBRN); and discrimination and bias. Among 27 mitigations identified through a literature review, we find that a broad range of risk mitigation measures are perceived as effective in reducing various systemic risks and technically feasible by domain experts. In particular, three mitigation measures stand out: safety incident reports and security information sharing, third-party pre-deployment model audits, and pre-deployment risk assessments. These measures show both the highest expert agreement ratings (>60\%) across all four risk areas and are most frequently selected in experts' preferred combinations of measures (>40\%). The surveyed experts highlighted that external scrutiny, proactive evaluation and transparency are key principles for effective mitigation of systemic risks. We provide policy recommendations for implementing the most promising measures, incorporating the qualitative contributions from experts. These insights should inform regulatory frameworks and industry practices for mitigating the systemic risks associated with general-purpose AI.
Operate and Monitor
Running, maintaining, and monitoring the AI system post-deployment
Developer
Entity that creates, trains, or modifies the AI system
Manage
Prioritising, responding to, and mitigating AI risks
