Claude Is Like a Smart but Unreliable Assistant
GaryMarcus · x · 2026-07-17
The author shared feedback on using Claude: while it significantly accelerates "low-value but time-consuming" tasks, it is highly error-prone and requires constant human supervision and constraint.
The core takeaway is that such models act more like a "smart but unreliable assistant." Failing to understand this makes it easy to be misled by the model's capabilities.
More from Models
- Gary Marcus says LLM math skills are like knowing only a car’s engine size — GaryMarcus · 2026-07-22
- OpenAI’s Codex + GPT-5.6 Sol hits 99% recall in Project APE verification tests — soumitrashukla9 · 2026-07-22
- OpenAI rolls out voice in GPT-Live, but the UI obscures search and reasoning — Graham_dePenros · 2026-07-22
- Moonshot’s Kimi K3 sets a new open-weights ECI record at 156 — scaling01 · 2026-07-22
- Nanbeige4.2-3B launches as a 3B Looped Transformer model that beats larger baselines — Wooden-Deer-1276 · 2026-07-22
- A post says six companies now beat Google’s best LLM, including two open-source models — soham_btw · 2026-07-22