Agents
not wrong!
In a recent Twitter exchange, Luca Ambrogioni stated that large‑language models (LLMs) likely possess fundamental limitations that may be obscured by progress in reasoning and agentic pipelines, thoug
In a recent Twitter exchange, Luca Ambrogioni stated that large‑language models (LLMs) likely possess fundamental limitations that may be obscured by progress in reasoning and agentic pipelines, though he admits the possibility of being wrong. Gary Marcus responded with “not wrong,” indicating agreement with Ambrogioni’s assessment. The discussion reflects ongoing debate about LLM intrinsic constraints versus engineering advances.
Related
- Real-Time Progress Prediction in Reasoning Language Models
- The Consistency Dilemma in LLMs: Generator-Evaluator Agreement and Vulnerability to Mistakes
- Self-signals Driven Multi-LLM Debate for Efficient and Accurate Reasoning
Source: Gary Marcus (X) | 2026-08-09