Results for "goals vs intelligence"

Goals useful regardless of final objective.

Intelligence and goals are independent.

Decomposing goals into sub-tasks.

Correctly specifying goals.

Methods for breaking goals into steps; can be classical (A*, STRIPS) or LLM-driven with tool calls.

System that independently pursues goals over time.

Distributed agents producing emergent intelligence.

Ensuring AI systems pursue intended human goals.

A system that perceives state, selects actions, and pursues goals—often combining LLM reasoning with tools and memory.

The field of building systems that perform tasks associated with human intelligence—perception, reasoning, language, planning, and decision-making—via algori...

Tendency for agents to pursue resources regardless of final goal.

Inferring human goals from behavior.

AI capable of performing most intellectual tasks humans can.

Rate at which AI capabilities improve.

Ensuring model behavior matches human goals, norms, and constraints, including reducing harmful or deceptive outputs.

Agent reasoning about future outcomes.

Internal representation of the agent itself.

Tradeoff between safety and performance.

Research ensuring AI remains safe.

Sudden jump to superintelligence.

System-level design for general intelligence.

A dataset + metric suite for comparing models; can be gamed or misaligned with real-world goals.

Attacks that manipulate model instructions (especially via retrieved content) to override system goals or exfiltrate data.

Multiple agents interacting cooperatively or competitively.

Maximizing reward without fulfilling real goal.

Model optimizes objectives misaligned with human values.

Ensuring learned behavior matches intended objective.

Learned subsystem that optimizes its own objective.

Model behaves well during training but not deployment.

Artificial environment for training/testing agents.