告诉HN：我对《神话》并不感到兴奋，对卡尔帕西也感到失望。
* 高层次原因：<p>这加大了贫富之间的差距。你认为独立开发者和个人有资源按令牌付费吗？Anthropic所走的方向只会导致更多的不平等，这次是小公司和大公司之间的智能不平等。<p>Anthropic在进行AI研究时故意削弱了模型。而且，Karpathy（一个叛逃到Anthropic的AI研究员）竟然有胆量赞美这个模型？<p>这些决定在美国政府要求下架模型之前就已经由Anthropic做出（这是一家积极推动AI监管和禁止开源模型的公司），所以他们不能将此归咎于政治因素。<p>* 现实原因：<p>我们原本计划在6月9日至6月23日（15天）期间实现100%的使用率。<p>结果我们只获得了3天的100%使用率和7天的50%使用率。<p>而且在第一天就被削弱了：<p>> ... 我们正在重新部署模型，使用一套新的分类器来针对和阻止更多的网络安全任务。在短期内，一些常规任务，如编码和调试，将回退到Opus 4.8。<p>如果我想要Fable级别的模型，已经有GPT-5.5-pro（出于某种原因在Codex中无法访问），我可以用我的订阅来使用。关于Fable的内容没有任何。
查看原文
* High level reasons:<p>It increases the gap between the rich and the poor. You think indie devs and individuals have the resources to pay per token? The direction Anthropic is taking is one that leads to more inequality, only this time it's intelligence inequality between small and large corps.<p>Anthropic also intentionally crippled the model when it comes to doing AI research. And then Karpathy (an AI researcher who defected to Anthropic) has the audacity to praise the model?<p>These decisions were already made by Anthropic (a company that actively wants AI regulation and banning of open source models) BEFORE the USG request to take down the model, so they can't blame the politics for this.<p>* Earthly reasons:<p>We were originally going to get 100% usage from June 9th to June 23rd (15 days)<p>Instead we got 100% usage for 3 days and 50% usage for 7.<p>And it's nerfed on day 1:<p>> ... we're redeploying the model with a new set of classifiers to target and block more cybersecurity tasks. In the near term, some routine tasks like coding and debugging will fall back to Opus 4.8<p>If I want Fable-class of models, there's already GPT-5.5-pro (not accessible in Codex, for some reason) which I can use with my subscription. Nothing about Fable