MIT AI Risk Repository · Risk Sub-Category · 73.01.02
Natural Language Underspecifies Goals
Category: Agentic LLMs Pose Novel Risks
Description
"For LLM-agents, both the goal and environment observations are typically specified in the prompt through natural language. While natural language may provide a richer and more natural means of specifying goals than alternatives such as hand-engineering objective functions, natural language still suffers from underspecification (Grice, 1975; Piantadosi et al., 2012). Furthermore, in practice, users may neglect fully specifying their goals, especially the information pertaining to elements of the environment that ought not to be changed (the classic frame problem (Shanahan, 2016)). Such undersp
From Foundational Challenges in Assuring Alignment and Safety of Large Language Models (Anwar2024), as extracted by the MIT AI Risk Repository (CC BY 4.0).
Classification
- Causal entity
- Other
- Intent
- Unintentional
- Timing
- Pre-deployment
Subdomain definition: AI systems acting in conflict with human goals or values, especially the goals of designers or users, or ethical standards. These misaligned behaviors may be introduced by humans during design and development, such as through reward hacking and goal misgeneralisation, or may result from AI using dangerous capabilities such as manipulation, deception, situational awareness to seek power, self-proliferate, or achieve other goals.
Real-world incidents in this subdomain
- Reinforcement Learning Reward Functions in Video Games
- Predictive Policing Program by Florida Sheriff’s Office Allegedly Violated Residents’ Rights and Targeted Children of Vulnerable Groups
- Image Classification of Battle Tanks
How other frameworks describe this risk
- Loss of control
- Loss of control
- Sudden loss of control
- AI leads to humans losing control of the future
- Risks from delegating decision-making power to misaligned AIs
- Risks from AIs developing goals and values that are different from humans
- Future AI systems might actively reduce human control
- Value specification
Other entries from Anwar2024
- Agentic LLMs Pose Novel Risks
- Goal-Directedness Incentivizes Undesirable Behaviors
- Safety Risks from Affordances Provided to LLM-agents
- Multi-Agent Safety Is Not Assured by Single-Agent Safety
- Foundationality May Cause Correlated Failures
- Groups of LLM-Agents May Show Emergent Functionality
- Collusion between LLM-Agents
- Dual-Use Capabilities Enable Malicious Use and Misuse of LLMs
- Misinformation and Manipulation
- Cybersecurity
- Cybersecurity
- Cybersecurity