📰 BloombergBloombergCodex · 实盘不做把握 6/10
北京时间·🕒 发布 10:53:33时区推断·⟳ 抓取 11:04:24·⚡ 决策 08-06 16:44:58 +5h40m·82.0 天前
DeepSeek 拟上调 AI 服务费用，但未披露具体区间，且影响停留在服务商竞争叙事
事件为未上市 DeepSeek 的定价调整，且仅给出“将显著上调”且未披露具体幅度，难以转化为可量化交易前提。
📄 原文Bloomberg↗ 原文
DeepSeek plans to implement a significant price increase across its AI services, an unusual shift from the Chinese tech…
不做 百度（BIDU.US）
无可执行的同向催化链：主体 DeepSeek 未上市，且对 Baidu 等受冲击企业仅有市场叙事层面提及，无明确金额、订单、营收占比或增量利润证据可支撑同向交易。
展开详情(成交明细 / 研究档案 / 原文全文 / 决策记录链)
研究档案
🔎 新鲜度:最早出现 2026-08-05 · 扩散 low · price-in 风险 medium ↗
消息传播链
- Bloomberg 披露 DeepSeek 通知用户计划对 AI 服务做明显提价，并给出 DeepSeek 与部分竞争方的当前价格对比，后续仍未见一手官方更早公告时间点。
关键理由
- 事件为未上市 DeepSeek 的定价调整，且仅给出“将显著上调”且未披露具体幅度，难以转化为可量化交易前提。
- 文本中主要是行业格局叙事，且缺少可核验的上市公司级订单/营收/利润增量链路，不能形成清晰同向交易候选。
主要风险
- 若后续仅靠行业情绪解读而非公开财务变化执行，会把“平台级竞争叙事”误当成可交易催化。
- 未披露提价细则与生效节奏，若后续仅为公关层解释且并未改变实际付费结构，容易出现追涨或先入场后失真风险。
原文全文
展开原文 · 2,268 字
DeepSeek plans to implement a significant price increase across its AI services, an unusual shift from the Chinese tech sector that has put pressure on US competitors with its aggressive pricing.
The Hangzhou-based company in recent days has led the conversation around the cost of high-performing artificial intelligence tools, as its V4 Flash showed itself capable of competing with the world’s best while costing cents where US competitors charged several dollars. Its pricing has been so far out of the usual range for AI models that it has prompted the idea of a DeepSeek danger zone, where costlier or less capable models would be obviated.
In a notice to users on Thursday, DeepSeek did not specify the exact increases, other than to call them substantial and urge users to plan accordingly. The company currently charges $0.14 per million input tokens and $0.28 per million output tokens — measures of work an AI model does to digest information and produce results — whereas Chinese rival Moonshot prices its Kimi K3 at $3 and $15 per million of each token type. Anthropic’s state-of-the-art Fable 5 service sets that pricing at $10 and $50, respectively.
Read More: China’s AI Blitz Creates ‘Death Zone’ for Rival US Model Makers
DeepSeek intends to build a massive data center in Inner Mongolia, part of a plan to secure 1 gigawatt of AI computing power that may include leasing additional capacity from other companies, Bloomberg News reported last week. It needs to build up the funding to support that kind of costly venture. Nvidia Chief Executive Officer Jensen Huang has previously estimated that a 1GW data center equipped with the most cutting-edge AI accelerators would cost about $50 billion, although costs in China are typically lower than in the US.
What Bloomberg Intelligence SaysChinese consumer interest in AI tools should remain buoyant through 2H26, as DeepSeek and ByteDance remained dominant in the country’s AI chatbot sector in June, even after high-profile model launches from Zhipu and Moonshot AI. The continued concentration around the two market leaders will likely place further pressure on sub-scale firms including Baidu, in a commoditized sector flooded with free-to-access tools.— Robert Lea, BI senior industry analyst
The Hangzhou-based company in recent days has led the conversation around the cost of high-performing artificial intelligence tools, as its V4 Flash showed itself capable of competing with the world’s best while costing cents where US competitors charged several dollars. Its pricing has been so far out of the usual range for AI models that it has prompted the idea of a DeepSeek danger zone, where costlier or less capable models would be obviated.
In a notice to users on Thursday, DeepSeek did not specify the exact increases, other than to call them substantial and urge users to plan accordingly. The company currently charges $0.14 per million input tokens and $0.28 per million output tokens — measures of work an AI model does to digest information and produce results — whereas Chinese rival Moonshot prices its Kimi K3 at $3 and $15 per million of each token type. Anthropic’s state-of-the-art Fable 5 service sets that pricing at $10 and $50, respectively.
Read More: China’s AI Blitz Creates ‘Death Zone’ for Rival US Model Makers
DeepSeek intends to build a massive data center in Inner Mongolia, part of a plan to secure 1 gigawatt of AI computing power that may include leasing additional capacity from other companies, Bloomberg News reported last week. It needs to build up the funding to support that kind of costly venture. Nvidia Chief Executive Officer Jensen Huang has previously estimated that a 1GW data center equipped with the most cutting-edge AI accelerators would cost about $50 billion, although costs in China are typically lower than in the US.
What Bloomberg Intelligence SaysChinese consumer interest in AI tools should remain buoyant through 2H26, as DeepSeek and ByteDance remained dominant in the country’s AI chatbot sector in June, even after high-profile model launches from Zhipu and Moonshot AI. The continued concentration around the two market leaders will likely place further pressure on sub-scale firms including Baidu, in a commoditized sector flooded with free-to-access tools.— Robert Lea, BI senior industry analyst
决策记录链 · 只追加
每次领取产生一条记录,重新决策追加一条并以 supersedes 指向被替代的记录(ADR 0014)。
这条决策早于决策账本(2026-10-08 起逐条追加),没有追加记录。