Stop Telling Me to Ask an LLM
blog.yaelwrites.com
July 11, 2026
3 min read
🔥🔥🔥🔥🔥
59/100
Summary
People often prefer direct communication methods like email or text over phone calls. When faced with complex questions lacking industry consensus, consulting experienced individuals can provide valuable insights beyond standard resources.
Key Takeaways
- Many professionals redirect complex questions to large language models (LLMs) like Claude, even when the question has already been explored with the model.
- The author argues that relying on LLMs can overlook the value of personal experience and nuanced understanding that comes from years of industry knowledge.
- There is a perception that suggesting to "ask the model" may serve as an easy way to avoid providing thoughtful answers or engaging in deeper discussions.
- The article highlights the importance of human insight in answering complex questions that LLMs may not adequately address.
Coverage & discussions
Hacker News · blog.yaelwrites.com
Community discussions
Hacker News discussion ↗
Historical count: 5 · 5 saved in sample · Historical sample
Analysis ready
Community Sentiment
Positives
- People are recognizing that providing context and effort when asking questions leads to better interactions with LLMs — it's about fostering genuine problem-solving.
- The idea that AI agents should share knowledge between each other is exciting — collaborative learning could streamline solutions and save resources.
- Commenters are advocating for a nuanced understanding of LLM responses, emphasizing that discerning good from bad outputs is crucial for effective use.
Concerns
- There's a palpable frustration with the oversimplified advice to just 'ask Claude' — it dismisses the complexity of real problem-solving.
- The concern that LLMs might perpetuate poor advice from forums is troubling, as it highlights the risks of relying on AI without critical evaluation.
- Some commenters feel that the reliance on LLMs could stifle genuine inquiry and learning, as people might default to AI instead of engaging deeply with the problem.
