MIT AI Risk Repository · Risk Sub-Category · 24.01.02
Difficult to develop metrics for evaluating benefits or harms caused by AI assistants
Category: Capability failures
Description
"Another difficulty facing AI assistant systems is that it is challenging to develop metrics for evaluating particular aspects of benefits or harms caused by the assistant – especially in a sufficiently expansive sense, which could involve much of society (see Chapter 19). Having these metrics is useful both for assessing the risk of harm from the system and for using the metric as a training signal."
From The Ethics of Advanced AI Assistants (Gabriel2024), as extracted by the MIT AI Risk Repository (CC BY 4.0).
Classification
- Subdomain
- 6.5 Governance failure
- Causal entity
- AI
- Intent
- Unintentional
- Timing
- Pre-deployment
Subdomain definition: Inadequate regulatory frameworks and oversight mechanisms failing to keep pace with AI development, leading to ineffective governance and the inability to manage AI risks appropriately.
Real-world incidents in this subdomain
- Nippon Life Alleged ChatGPT Practiced Law Without a License in Illinois Disability Case
- OpenAI Allegedly Did Not Alert RCMP After ChatGPT Flagged Violent Chats Before British Columbia School Shooting
- Remotely Operated Taser-Armed Drones Proposed by Taser Manufacturer as Defense for School Shootings in the US
How other frameworks describe this risk
- Liability issues in case of accidents
- Mobility
- Faster scientific progress makes it harder for governance to keep pace with development
- Type 1: Diffusion of responsibility
- Products Liability Law
- Benchmarking (Cross-lingual data contamination)
- Benchmarking (Guideline contamination)
- Benchmarking (Annotation contamination)
Other entries from Gabriel2024
- Capability failures
- Lack of capability for task
- Safe exploration problem with widely deployed AI assistants
- Goal-related failures
- Misaligned consequentialist reasoning
- Specification gaming
- Goal misgeneralisation
- Deceptive alignment
- Malicious Uses
- Offensive Cyber Operations (General)
- AI-Powered Spear-Phishing at Scale
- AI-Assisted Software Vulnerability Discovery