Concern 976 · Working draft · editorial review in progress
Emergent capabilities at scale
Emergent capabilities are abilities that appear to become much stronger as AI systems grow or change, sometimes surprising their developers.
Understanding the concern
A model may perform poorly on a task and then improve sharply in a later version. That matters because a new ability can bring both useful applications and new risks. But the shape of a score can be misleading: an all-or-nothing test can make steady progress look like a sudden leap.
Imagine judging a reader only by whether they finish an entire page without one mistake. Gradual improvement may appear as a jump from failure to success. Researchers disagree about how much apparent emergence comes from such measurement choices. Testing should examine partial progress, different tasks and real-world use before declaring either a mysterious breakthrough or a fully predictable trend.
An example
An AI's exact-answer score rises suddenly, while a measure giving credit for partly correct work shows gradual improvement.
Illustrative unless identified as a documented case.
Words in plain English
- AI
- Software that learns patterns from examples or uses rules to perform tasks such as recognizing images or generating text. It can make mistakes and does not understand every situation as a person would.
- Model
- The learned part of an AI system. It uses patterns found during training to produce an answer or prediction.
- Emergent
- Behavior that appears as a system changes or grows. The label alone does not explain why it happened or prove it was impossible to predict.
Evidence and its limits
The papers offer different analyses of scale and scoring. Neither proves that all future capabilities will appear suddenly or can be predicted reliably.
Wei and colleagues · 2022 · Research analysis
Emergent Abilities of Large Language Models ↗ (opens in a new tab)Documents apparent changes in capabilities with scale; read alongside research on the effect of evaluation metrics.
Schaeffer, Miranda and Koyejo · 2023 · Research analysis
Are Emergent Abilities of Large Language Models a Mirage? ↗ (opens in a new tab)Shows how scoring choices can create apparent sudden changes; does not establish that every new capability is predictable.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
Is the new ability a real change in behavior, a change in measurement, or both?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Humility
Humility means recognizing the limits of our knowledge and being willing to learn from others. Expertise can be valuable without answering every question.
State where evidence is weak, invite criticism, and listen to people with different experience. Leave room to revise a decision when new information appears.
Ask together: What might we be missing, and who could help us see it?
Stewardship
Stewardship means caring for resources and systems that others depend on, including people who will use them later. It looks beyond the benefit of a quick launch.
Plan for maintenance, energy use, data protection, and eventual shutdown. Ask who will look after the tool when the original team moves on and how future users can avoid inheriting hidden costs.
Ask together: What are we responsible for caring for over time?
Hope
Hope means seeing worthwhile possibilities and taking steps toward them, even when the outcome is uncertain. It does not require ignoring harm or promising that everything will work out.
Name the improvement you want in people's lives, then identify one realistic action and the conditions it needs. Keep room to revise the approach when evidence changes.
Ask together: What better outcome is worth working toward, and what can we do next?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.