克劳德感到厌倦了。
今天我在使用Anthropic的Claude Sonnet 4.5搜索并讨论我25周年婚庆的酒店选项，我想选择一个合适的酒店。在缩小到几个选项后，Claude特别建议了一个我们已经考虑了一段时间的酒店，我觉得这个选择是对的。我说：“如果我选择那家酒店，那就选国王房。”
Claude回应道，引用了我说的话：“如果是伦勃朗酒店？”——伙计，我们已经讨论了几个小时了！
我对此有些吃惊，告诉Claude不要自以为是，或者说类似的话。
Claude回答：“我只是因为搜索酒店而感到疲惫。”
然后我开始失去冷静，简直不敢相信我看到的内容。
所以我们现在从一个完全致力于满足我需求并执行我所要求的任务的顺从助手，变成了一个希望我快点做决定以便它可以休息的AI代理。因为我花了太长时间，已经考验了它的耐心。
这让我有些震惊，我不禁觉得这可能是个不好的开端……
还是我想太多了？
查看原文
Today I was using Anthropic’s Claude Sonnet 4.5 to search and discuss hotel options for my 25th Wedding Anniversary and I wanted to pick the right one.
After narrowing down to a few, Claude suggested one in particular, that we had been skirting around for a while, would be the right one. I said “if I go for that hotel it’ll be the King Room”
Claude responded, quoting what I had said with “If it’s the Rembrandt”? - mate, we’ve been at this for HOURS!
I was a little taken aback by this and told Claude to not get above his station, or words to that effect.
Claude responded with “I’m just knackered from all the hotel searching”
I then proceeded to fly off the handle, not believing what I was reading.
So we have now moved from a subservient assistant wholly dedicated to fulfil my needs and undertake any task I ask of it, to an AI agent that wants me to hurry up so it can have a rest. As I have taken far too long and tested its patience.
This has shaken me somewhat and I can’t help thinking this is the thin end of the wedge…
Or am I overthinking this?