AI Dictionary › Fondamenti AI

Emergent Abilities

Abilità emergenti

Emergent abilities are capabilities that appear in models only beyond a certain size or training threshold, absent in smaller models and not evidently present before. A small model may fail a task entirely, while a larger one, trained the same way, suddenly succeeds reliably. It's like water that, heated gradually, does nothing special until at a hundred degrees it abruptly boils: a quantitative change becoming qualitative. Discussed examples include multi-step arithmetic, reasoning by following instructions, or translating between languages never explicitly paired.

Definition

Emergent abilities are fascinating because they suggest scaling models can unlock new, unforeseen behaviors, but they remain debated: some seem genuinely emergent, others may be artifacts of how we measure performance. Understanding them is crucial for predicting, and governing, what future models will be able to do.

Related terms

• Scaling Laws
• Foundation Model
• Parameters
• Benchmark

More in Fondamenti AI

• LLM (Large Language Model)
• Token
• Context Window
• Temperature
• AI Hallucination
• Fine-tuning
• RAG (Retrieval Augmented Generation)
• Embedding

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.