{"id":357,"date":"2026-04-27T07:20:57","date_gmt":"2026-04-26T23:20:57","guid":{"rendered":"http:\/\/www.faiyi.com\/?p=357"},"modified":"2026-04-27T12:36:44","modified_gmt":"2026-04-27T04:36:44","slug":"ai%e5%8a%a8%e6%80%81%e6%af%8f%e6%97%a5%e7%ae%80%e6%8a%a5-2026-04-27","status":"publish","type":"post","link":"http:\/\/www.faiyi.com\/?p=357","title":{"rendered":"AI\u52a8\u6001\u6bcf\u65e5\u7b80\u62a5 2026-04-27"},"content":{"rendered":"

\u65e5\u671f\uff1a2026-04-27<\/p>\n

\u672c\u671f\u805a\u7126\uff1a\u91cd\u70b9\u5173\u6ce8\u6a21\u578b\u53d1\u5e03\u4e0e release notes\u3001\u5b98\u65b9 engineering blog\u3001AI coding \/ agent \/ SRE\u3001\u8bc4\u6d4b\u699c\u5355\u53d8\u5316\u3001\u5f00\u53d1\u8005\u5b9e\u8df5\u535a\u5ba2\u3001\u6846\u67b6\u751f\u6001\u3001\u5f00\u6e90\u6a21\u578b\u4e0e\u771f\u5b9e\u7528\u6237\u89c6\u89d2\uff1b\u5f53 HN\u3001Reddit\u3001Hugging Face \u7b49\u793e\u533a\u6e90\u53ef\u8bbf\u95ee\u65f6\u4f18\u5148\u7eb3\u5165\u3002<\/p>\n

\n
    \n
  1. \n

    Artificial Analysis \u6700\u65b0\u6a21\u578b\u6392\u540d\u89c2\u5bdf<\/strong>\uff08Artificial Analysis\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>Artificial Analysis \u6700\u65b0\u6a21\u578b\u6392\u540d\u663e\u793a\uff0cGPT-5.5 (xhigh) \u4ee5 60 \u5206\u4f4d\u5c45\u667a\u80fd\u6307\u6570\u699c\u9996\uff0cGPT-5.5 (high) \u4ee5 59 \u5206\u7d27\u968f\u5176\u540e\uff0cClaude Opus 4.7 (Max Effort) \u4e0e Gemini 3.1 Pro Preview \u5e76\u5217\u7b2c\u4e09\uff0857 \u5206\uff09\u3002\u5728\u8f93\u51fa\u901f\u5ea6\u65b9\u9762\uff0cMercury 2 \u4ee5\u6bcf\u79d2 687 \u4e2a token \u9886\u5148\uff0cGranite 3.3 8B \u4ee5 333 t\/s \u4f4d\u5c45\u7b2c\u4e8c\u3002\u5ef6\u8fdf\u6700\u4f4e\u7684\u662f Ministral 3 3B\uff080.45 \u79d2\uff09\u548c LFM2 24B A2B\uff080.50 \u79d2\uff09\u3002\u8be5\u5e73\u53f0\u76ee\u524d\u5171\u8bc4\u4f30\u4e86 361 \u4e2a\u6a21\u578b\uff0c\u63d0\u4f9b\u7aef\u5230\u7aef\u54cd\u5e94\u65f6\u95f4\u3001\u63a8\u7406\u6a21\u578b\u601d\u8003\u65f6\u95f4\u7b49\u8be6\u7ec6\u6307\u6807\u5bf9\u6bd4\u3002<\/p>\n

    English Summary:<\/strong> Artificial Analysis' latest model rankings show GPT-5.5 (xhigh) leading the Intelligence Index with a score of 60, followed by GPT-5.5 (high) at 59, and Claude Opus 4.7 (Max Effort) tied with Gemini 3.1 Pro Preview at 57. For output speed, Mercury 2 leads at 687 tokens\/s, with Granite 3.3 8B at 333 t\/s. Lowest latency models are Ministral 3 3B (0.45s) and LFM2 24B A2B (0.50s). The platform evaluates 361 models total, providing detailed metrics including end-to-end response time and reasoning model thinking time.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  2. \n

    Introducing Claude Opus 4.7<\/strong>\uff08Anthropic News\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>Anthropic \u53d1\u5e03 Claude Opus 4.7\uff0c\u5728\u591a\u9879\u57fa\u51c6\u6d4b\u8bd5\u4e2d\u8868\u73b0\u663e\u8457\u63d0\u5347\u3002\u5728 Databricks OfficeQA Pro \u6587\u6863\u63a8\u7406\u4efb\u52a1\u4e2d\uff0cOpus 4.7 \u6bd4 4.6 \u7248\u672c\u51cf\u5c11 21% \u7684\u9519\u8bef\u7387\uff0c\u6210\u4e3a\u4f01\u4e1a\u6587\u6863\u5206\u6790\u7684\u6700\u4f73 Claude \u6a21\u578b\u3002\u5728 Rakuten-SWE-Bench \u4e0a\uff0c\u5176\u89e3\u51b3\u751f\u4ea7\u4efb\u52a1\u7684\u80fd\u529b\u662f 4.6 \u7684\u4e09\u500d\uff0c\u4ee3\u7801\u8d28\u91cf\u548c\u6d4b\u8bd5\u8d28\u91cf\u5747\u6709\u53cc\u4f4d\u6570\u63d0\u5347\u3002Ramp \u53cd\u9988\u79f0\u8be5\u7248\u672c\u5728\u4ee3\u7406\u56e2\u961f\u534f\u4f5c\u4e2d\u89d2\u8272\u4fdd\u771f\u5ea6\u3001\u6307\u4ee4\u9075\u5faa\u548c\u590d\u6742\u63a8\u7406\u80fd\u529b\u66f4\u5f3a\u3002Hebbia \u5219\u89c2\u5bdf\u5230\u5de5\u5177\u8c03\u7528\u51c6\u786e\u6027\u548c\u89c4\u5212\u80fd\u529b\u6709\u53cc\u4f4d\u6570\u63d0\u5347\u3002<\/p>\n

    English Summary:<\/strong> Anthropic released Claude Opus 4.7 with significant improvements across benchmarks. On Databricks' OfficeQA Pro, it shows 21% fewer errors than Opus 4.6 in document reasoning, becoming the best-performing Claude model for enterprise document analysis. On Rakuten-SWE-Bench, it resolves 3x more production tasks than Opus 4.6 with double-digit gains in Code and Test Quality. Ramp reports stronger role fidelity and complex reasoning in agent-team workflows, while Hebbia observed double-digit jumps in tool call accuracy and planning.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  3. \n

    Featured An update on recent Claude Code quality reports<\/strong>\uff08Anthropic Engineering\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>Anthropic \u5de5\u7a0b\u56e2\u961f\u53d1\u5e03\u5173\u4e8e Claude Code \u8fd1\u671f\u8d28\u91cf\u95ee\u9898\u7684\u590d\u76d8\u62a5\u544a\u3002\u4e3b\u8981\u95ee\u9898\u5305\u62ec\uff1a4 \u6708\u521d\u4ee3\u7801\u5ba1\u67e5\u529f\u80fd\u9057\u6f0f\u5173\u952e bug\uff0c\u539f\u56e0\u662f\u7cfb\u7edf\u63d0\u793a\u8bcd\u53d8\u66f4\u9650\u5236\u4e86\u8f93\u51fa\u957f\u5ea6\uff08\u8981\u6c42\u5de5\u5177\u8c03\u7528\u95f4\u6587\u672c\u4e0d\u8d85\u8fc7 25 \u8bcd\uff0c\u6700\u7ec8\u56de\u590d\u4e0d\u8d85\u8fc7 100 \u8bcd\uff09\uff0c\u5bfc\u81f4\u6a21\u578b\u667a\u80fd\u4e0b\u964d\uff1b\u4ee5\u53ca\u7f13\u5b58\u4f18\u5316\u9519\u8bef\u5730\u4e22\u5f03\u4e86\u63a8\u7406\u5386\u53f2\u3002\u56e2\u961f\u5df2\u56de\u6eda\u76f8\u5173\u53d8\u66f4\uff0c\u5c06 Opus 4.7 \u9ed8\u8ba4 effort \u7ea7\u522b\u6062\u590d\u4e3a xhigh\uff0c\u5e76\u4fee\u590d\u4e86\u7f13\u5b58\u95ee\u9898\u3002\u6d4b\u8bd5\u663e\u793a\uff0c\u5728\u5b8c\u6574\u4ee3\u7801\u4ed3\u5e93\u4e0a\u4e0b\u6587\u4e2d\uff0cOpus 4.7 \u80fd\u591f\u53d1\u73b0 Opus 4.6 \u9057\u6f0f\u7684 bug\u3002<\/p>\n

    English Summary:<\/strong> Anthropic Engineering published a postmortem on recent Claude Code quality issues. Key problems included: a system prompt change limiting output length (\u226425 words between tool calls, \u2264100 words final responses) that reduced model intelligence, causing code review to miss critical bugs in early April; and a caching optimization that incorrectly dropped reasoning history. The team has rolled back these changes, restored xhigh as the default effort level for Opus 4.7, and fixed the caching issue. Testing showed Opus 4.7 could catch bugs that Opus 4.6 missed when given complete repository context.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  4. \n

    Scaling Managed Agents: Decoupling the brain from the hands<\/strong>\uff08Anthropic Engineering\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>Anthropic \u5de5\u7a0b\u535a\u5ba2\u53d1\u5e03\u5173\u4e8e\u6258\u7ba1\u4ee3\u7406\uff08Managed Agents\uff09\u7684\u67b6\u6784\u6587\u7ae0\uff0c\u63d0\u51fa\u5c06\u4ee3\u7406\u7684"\u5927\u8111"\u4e0e"\u6267\u884c\u5668"\u89e3\u8026\u7684\u8bbe\u8ba1\u7406\u5ff5\u3002\u8be5\u7cfb\u7edf\u901a\u8fc7\u865a\u62df\u5316\u4e09\u4e2a\u6838\u5fc3\u7ec4\u4ef6\u2014\u2014\u4f1a\u8bdd\uff08append-only \u65e5\u5fd7\uff09\u3001 harness\uff08\u8c03\u7528 Claude \u5e76\u8def\u7531\u5de5\u5177\u8c03\u7528\u7684\u5faa\u73af\uff09\u548c\u6c99\u7bb1\uff08\u4ee3\u7801\u6267\u884c\u73af\u5883\uff09\u2014\u2014\u5b9e\u73b0\u7075\u6d3b\u7ec4\u5408\u3002\u8fd9\u79cd\u5143 harness \u67b6\u6784\u4e0d\u9884\u8bbe\u5177\u4f53\u5b9e\u73b0\uff0c\u5141\u8bb8\u6839\u636e\u4efb\u52a1\u9700\u6c42\u5207\u6362\u4e0d\u540c harness\uff08\u5982 Claude Code \u6216\u7279\u5b9a\u9886\u57df\u4ee3\u7406\uff09\uff0c\u4f7f\u7cfb\u7edf\u80fd\u968f\u6a21\u578b\u80fd\u529b\u63d0\u5347\u800c\u6f14\u8fdb\u3002\u6587\u7ae0\u5f3a\u8c03\u907f\u514d"\u517b\u5ba0\u7269"\u5f0f\u7684\u57fa\u7840\u8bbe\u65bd\u7ed1\u5b9a\uff0c\u63d0\u5021\u901a\u8fc7\u6807\u51c6\u5316\u63a5\u53e3\u5b9e\u73b0\u7ec4\u4ef6\u53ef\u66ff\u6362\u6027\u3002<\/p>\n

    English Summary:<\/strong> Anthropic's Engineering Blog published an architecture article on Managed Agents, proposing decoupling the agent's "brain" from its "hands." The system virtualizes three core components\u2014session (append-only log), harness (loop calling Claude and routing tool calls), and sandbox (execution environment)\u2014enabling flexible composition.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  5. \n

    Our principles<\/strong>\uff08OpenAI News\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>OpenAI \u53d1\u5e03\u7531 Sam Altman \u64b0\u5199\u7684\u4e94\u9879\u6838\u5fc3\u539f\u5219\uff0c\u9610\u8ff0\u5176\u786e\u4fdd AGI \u9020\u798f\u5168\u4eba\u7c7b\u7684\u4f7f\u547d\u3002\u539f\u5219\u5305\u62ec\uff1a1\uff09\u6c11\u4e3b\u5316\u2014\u2014\u62b5\u5236\u6280\u672f\u5c06\u6743\u529b\u96c6\u4e2d\u4e8e\u5c11\u6570\u4eba\u7684\u8d8b\u52bf\uff0c\u786e\u4fdd\u5173\u952e\u51b3\u7b56\u901a\u8fc7\u6c11\u4e3b\u7a0b\u5e8f\u5236\u5b9a\uff1b2\uff09\u8d4b\u80fd\u2014\u2014\u76f8\u4fe1 AI \u80fd\u5e2e\u52a9\u6bcf\u4e2a\u4eba\u5b9e\u73b0\u76ee\u6807\u3001\u5b66\u4e60\u6210\u957f\uff0c\u901a\u8fc7\u666e\u53ca\u6613\u7528\u4e14\u8ba1\u7b97\u80fd\u529b\u5f3a\u5927\u7684 AI \u7cfb\u7edf\uff0c\u8ba9\u4eba\u4eec\u521b\u9020\u65b0\u4ef7\u503c\u5e76\u63d0\u5347\u751f\u6d3b\u8d28\u91cf\uff1b3\uff09\u666e\u904d\u7e41\u8363\u2014\u2014\u5e0c\u671b\u672a\u6765\u6bcf\u4e2a\u4eba\u90fd\u80fd\u62e5\u6709\u7f8e\u597d\u751f\u6d3b\uff0c\u653f\u5e9c\u53ef\u80fd\u9700\u8981\u8003\u8651\u65b0\u7684\u7ecf\u6d4e\u6a21\u5f0f\u4ee5\u786e\u4fdd\u4eba\u4eba\u53c2\u4e0e\u4ef7\u503c\u521b\u9020\uff0c\u540c\u65f6\u9700\u8981\u5efa\u8bbe\u5927\u91cf AI \u57fa\u7840\u8bbe\u65bd\u5e76\u964d\u4f4e\u6210\u672c\u3002<\/p>\n

    English Summary:<\/strong> OpenAI published five core principles written by Sam Altman outlining its mission to ensure AGI benefits all humanity. The principles include: 1) Democratization\u2014resisting technology's potential to consolidate power among the few, ensuring key AI decisions are made through democratic processes; 2) Empowerment\u2014believing AI can help everyone achieve goals and learn, putting easy-to-use AI with abundant compute into everyone's hands to create value and improve quality of life; 3) Universal Prosperity\u2014wanting a future where everyone can have an excellent life, with governments potentially needing new economic models to ensure broad participation in value creation and massive AI infrastructure buildout to drive costs down.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  6. \n

    To buy this Bay Area home, you\u2019ll need Anthropic equity<\/strong>\uff08TechCrunch AI\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>\u6e7e\u533a\u623f\u5730\u4ea7\u51fa\u73b0\u5947\u7279\u4ea4\u6613\uff1a\u6295\u8d44\u94f6\u884c\u5bb6Storm Duncan\u6b32\u4ee5\u4f4d\u4e8eMill Valley\u768413\u82f1\u4ea9\u8c6a\u5b85\u4ea4\u6362Anthropic\u80a1\u6743\u3002\u8be5\u623f\u4ea72019\u5e74\u4ee5475\u4e07\u7f8e\u5143\u8d2d\u5165\uff0c\u76ee\u524d\u7531\u4e00\u4f4d\u77e5\u540dVC\u79df\u4f4f\u3002Duncan\u79f0\u6b64\u4e3a"\u591a\u5143\u5316\u6295\u8d44"\u7b56\u7565\u2014\u2014\u4ed6\u770b\u597dAI\u672a\u6765\u4f46\u6301\u4ed3\u4e0d\u8db3\uff0c\u800c\u5e74\u8f7bAnthropic\u5458\u5de5\u53ef\u80fd\u6070\u597d\u76f8\u53cd\u3002\u4ea4\u6613\u91c7\u7528\u79c1\u4e0b\u80a1\u6743\u8f6c\u8ba9\u5f62\u5f0f\uff0c\u4e70\u65b9\u65e0\u9700\u51fa\u552e\u80a1\u7968\uff0c\u4f46\u9700\u5728\u9501\u5b9a\u671f\u5185\u8ba9\u6e2120%\u7684\u4e0a\u6da8\u6536\u76ca\u3002\u8fd9\u4e00\u4ea4\u6613\u6298\u5c04\u51faAI\u521d\u521b\u516c\u53f8\u80a1\u6743\u5728\u6e7e\u533a\u5df2\u6210\u4e3a\u4e00\u79cd"\u51c6\u8d27\u5e01"\u8d44\u4ea7\u3002<\/p>\n

    English Summary:<\/strong> A Bay Area investment banker is offering a 13-acre Mill Valley property in exchange for Anthropic equity rather than cash. Storm Duncan, who purchased the home for $4.75M in 2019, describes the swap as a "diversification play"\u2014he's underweight AI exposure while young Anthropic employees may be overweight. The private transaction allows buyers to retain their shares while transferring 20% of upside during the lockup period. The unusual deal highlights how AI startup equity has become a quasi-currency in Silicon Valley's high-end real estate market.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  7. \n

    [AINews] DeepSeek V4 Pro (1.6T-A49B) and Flash (284B-A13B), Base and Instruct \u2014 runnable on Huawei Ascend chips<\/strong>\uff08Latent Space\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>DeepSeek\u53d1\u5e03V4\u7cfb\u5217\u6a21\u578b\uff0c\u5305\u62ecPro\uff081.6T\u603b\u53c2\u6570\/49B\u6fc0\u6d3b\uff09\u548cFlash\uff08284B\/13B\uff09\u4e24\u4e2a\u7248\u672c\uff0c\u5747\u652f\u6301100\u4e07token\u4e0a\u4e0b\u6587\uff0c\u91c7\u7528MIT\u5f00\u6e90\u534f\u8bae\u3002\u6280\u672f\u4eae\u70b9\u5305\u62ec\uff1aCompressed Sparse Attention\u4e0eHeavily Compressed Attention\u673a\u5236\u4f7fKV\u7f13\u5b58\u8f83V3.2\u51cf\u5c11\u7ea610\u500d\uff1bFP4\/FP8\u6df7\u5408\u7cbe\u5ea6\u8bad\u7ec3\uff1b\u57fa\u4e8eMuon\u4f18\u5316\u5668\u3002\u8bc4\u6d4b\u663e\u793aV4 Pro\u5728Artificial Analysis Intelligence Index\u5f97\u520652\uff0c\u4f4d\u5217\u5f00\u6e90\u6a21\u578b\u7b2c\u4e8c\uff08\u4ec5\u6b21\u4e8eKimi K2.6\u768454\uff09\uff0c\u5728GDPval-AA agentic\u4efb\u52a1\u4e0a\u9886\u5148\u5f00\u6e90\u9635\u8425\u3002\u5b9a\u4ef7\u6781\u5177\u4fb5\u7565\u6027\uff1aFlash\u7248\u8f93\u5165\/\u8f93\u51fa\u4ec5$0.14\/$0.28\u6bcf\u767e\u4e07token\u3002\u8be5\u6a21\u578b\u540c\u65f6\u652f\u6301\u534e\u4e3aAscend\u82af\u7247\uff0c\u88ab\u89c6\u4e3a\u4e2d\u56fdAI\u81ea\u4e3b\u53ef\u63a7\u7684\u91cd\u8981\u91cc\u7a0b\u7891\u3002<\/p>\n

    English Summary:<\/strong> DeepSeek released the V4 family with Pro (1.6T\/49B active) and Flash (284B\/13B) variants, featuring 1M-token context and MIT licensing. Key innovations include Compressed Sparse Attention and Heavily Compressed Attention reducing KV cache by ~10x vs V3.2, FP4\/FP8 mixed training, and Muon optimizer. V4 Pro scores 52 on Artificial Analysis Intelligence Index (#2 open weights behind Kimi K2.6 at 54) and leads open models on GDPval-AA agentic benchmarks. Aggressive pricing at $0.14\/$0.28 per 1M tokens for Flash.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  8. \n

    Building Workforce AI Agents with Visier and Amazon Quick<\/strong>\uff08AWS ML Blog\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>AWS\u5b98\u65b9\u535a\u5ba2\u4ecb\u7ecdVisier Workforce AI\u5e73\u53f0\u4e0eAmazon Quick\u901a\u8fc7Model Context Protocol (MCP)\u7684\u96c6\u6210\u65b9\u6848\u3002\u8be5\u65b9\u6848\u4f7f\u77e5\u8bc6\u5de5\u4f5c\u8005\u80fd\u5728\u7edf\u4e00Agent\u5de5\u4f5c\u7a7a\u95f4\u4e2d\u67e5\u8be2\u4eba\u529b\u8d44\u6e90\u6570\u636e\u2014\u2014Visier\u63d0\u4f9b\u5b9e\u65f6\u5458\u5de5\u5206\u6790\u6570\u636e\uff08\u5982\u5728\u804c\u4eba\u6570\u3001\u5e73\u5747\u4efb\u671f\u3001\u9ad8\u7ee9\u6548\u5458\u5de5\u6bd4\u4f8b\uff09\uff0cAmazon Quick\u5219\u6574\u5408\u4f01\u4e1a\u5185\u90e8\u653f\u7b56\u6587\u6863\u4e0e\u9884\u7b97\u76ee\u6807\u3002\u6587\u7ae0\u6f14\u793a\u4e86HR\u4e0e\u8d22\u52a1\u89d2\u8272\u5982\u4f55\u534f\u4f5c\u51c6\u5907\u9886\u5bfc\u529b\u4f1a\u8bae\uff1a\u4ece\u81ea\u7136\u8bed\u8a00\u67e5\u8be2\u5230\u81ea\u52a8\u751f\u6210\u5305\u542b\u98ce\u9669\u8bc4\u4f30\u4e0e\u5efa\u8bae\u884c\u52a8\u7684\u7b80\u62a5\u3002Quick Flows\u529f\u80fd\u53ef\u5c06\u6b64\u6d41\u7a0b\u81ea\u52a8\u5316\uff0c\u6bcf\u5468\u5b9a\u65f6\u751f\u6210 workforce health score \u5e76\u63a8\u9001\u81f3\u90ae\u7bb1\u6216Slack\u3002<\/p>\n

    English Summary:<\/strong> AWS ML Blog details integrating Visier's Workforce AI platform with Amazon Quick via Model Context Protocol (MCP). The solution gives knowledge workers a unified agentic workspace to query HR data\u2014Visier supplies live workforce analytics (headcount, tenure, high-performer ratios) while Amazon Quick incorporates internal policy documents and budget targets. The post demonstrates HR and finance personas collaborating on leadership meeting prep: from natural language queries to auto-generated briefings with risk assessments and recommended actions.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  9. \n

    Presentation: Deepfakes, Disinformation, and AI Content are Taking over the Internet<\/strong>\uff08InfoQ AI\/ML\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>\u524dGoogle\u4fe1\u4efb\u4e0e\u5b89\u5168\u4ea7\u54c1\u8d1f\u8d23\u4eba\u3001Shape Security CTO Shuman Ghosemajumder\u5728QCon AI\u7684\u6f14\u8bb2\u63a2\u8ba8\u751f\u6210\u5f0fAI\u5982\u4f55\u4ece\u521b\u610f\u5de5\u5177\u6f14\u53d8\u4e3a\u5927\u89c4\u6a21\u865a\u5047\u4fe1\u606f\u4e0e\u6b3a\u8bc8\u6b66\u5668\u3002\u4ed6\u63d0\u51fa"\u865a\u5047\u4fe1\u606f\u81ea\u52a8\u5316"\u6846\u67b6\uff0c\u6307\u51faAI\u5185\u5bb9\u5df2\u8fdb\u5165YouTube Shorts\u548cTikTok\u9ed8\u8ba4\u63a8\u8350\u6d41\uff08\u4f30\u8ba1\u5360\u6bd420-30%\uff09\uff0c\u4e14\u771f\u5b9e\u5185\u5bb9\u7ecfAI\u6ee4\u955c\u5904\u7406\u540e\u66f4\u96be\u4e0e\u5047\u5185\u5bb9\u533a\u5206\u3002\u6f14\u8bb2\u5f3a\u8c03CAPTCHA\u5728AI\u65f6\u4ee3\u5df2\u5931\u6548\uff08\u673a\u5668\u8bc6\u522b\u738799.8% vs \u4eba\u7c7b33%\uff09\uff0c\u5e76\u5206\u4eab\u4e86\u9999\u6e2f\u67d0\u5de5\u7a0b\u516c\u53f8\u56e0\u6df1\u5ea6\u4f2a\u9020Zoom\u4f1a\u8bae\u88ab\u9a972500\u4e07\u7f8e\u5143\u7684\u6848\u4f8b\u3002\u4ed6\u5efa\u8bae\u4f01\u4e1a\u91c7\u7528\u96f6\u4fe1\u4efb"\u7f51\u7edc\u878d\u5408"\u7b56\u7565\uff0c\u7ed3\u5408\u884c\u4e3a\u5206\u6790\u4e0e\u591a\u56e0\u7d20\u8ba4\u8bc1\u5e94\u5bf9AI\u9a71\u52a8\u7684\u793e\u4f1a\u5de5\u7a0b\u653b\u51fb\u3002<\/p>\n

    English Summary:<\/strong> Shuman Ghosemajumder (ex-Google Trust & Safety, Shape Security CTO) presented at QCon AI on generative AI's evolution from creative tool to disinformation weapon. His "Disinformation Automation" framework notes AI content already comprises 20-30% of YouTube Shorts\/TikTok feeds, with real content AI-filtered to be indistinguishable. He highlights CAPTCHA's failure (99.8% machine vs 33% human solve rates) and a $25M Hong Kong deepfake Zoom scam.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  10. \n

    [AINews] GPT 5.5 and OpenAI Codex Superapp<\/strong>\uff08Latent Space\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>OpenAI\u53d1\u5e03GPT-5.5\u4e0eCodex\u91cd\u5927\u66f4\u65b0\u3002GPT-5.5\u5b9a\u4f4d\u4e3a"\u9762\u5411\u771f\u5b9e\u5de5\u4f5c\u7684\u65b0\u578b\u667a\u80fd"\uff0c\u5728Artificial Analysis Intelligence Index\u4e0a\u767b\u9876\uff0c\u4e14medium\u7248\u672c\u4ee5\u7ea61\/4\u6210\u672c\uff08$1,200 vs $4,800\uff09\u8fbe\u5230Claude Opus 4.7 max\u7684\u540c\u7b49\u667a\u80fd\u6c34\u5e73\u3002\u5173\u952e\u6307\u6807\u5305\u62ec\uff1aTerminal-Bench 2.0\u8fbe82.7%\u3001SWE-Bench Pro 58.6%\u3001GDPval 84.9%\u3002API\u5b9a\u4ef7\u4e3a$5\/$30\uff08\u6807\u51c6\u7248\uff09\u548c$30\/$180\uff08Pro\u7248\uff09\u6bcf\u767e\u4e07token\uff0c\u652f\u6301100\u4e07\u4e0a\u4e0b\u6587\u3002Codex\u540c\u6b65\u5347\u7ea7\uff0c\u65b0\u589e\u6d4f\u89c8\u5668\u63a7\u5236\u3001Sheets\/Slides\/Docs\u96c6\u6210\u3001OS\u7ea7\u542c\u5199\u548c\u81ea\u52a8\u5ba1\u6838\u6a21\u5f0f\uff0c\u7531"guardian"\u4ee3\u7406\u51cf\u5c11\u4eba\u5de5\u5ba1\u6279\u3002OpenAI\u6b63\u5c06Codex\u6253\u9020\u4e3a\u8d85\u7ea7\u5e94\u7528\u6218\u7565\u7684\u6838\u5fc3\u3002<\/p>\n

    English Summary:<\/strong> OpenAI launched GPT-5.5 and major Codex updates. GPT-5.5 is positioned as "a new class of intelligence for real work," topping Artificial Analysis Intelligence Index with its medium variant matching Claude Opus 4.7 max at roughly one-quarter cost ($1,200 vs $4,800). Key benchmarks: 82.7% Terminal-Bench 2.0, 58.6% SWE-Bench Pro, 84.9% GDPval. API pricing at $5\/$30 (standard) and $30\/$180 (Pro) per 1M tokens with 1M context. Codex gained browser control, Sheets\/Slides\/Docs integration, OS-wide dictation, and Auto-review mode using a "guardian" agent to reduce approvals. OpenAI is positioning Codex as the foundation of its superapp strategy.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  11. \n

    GPT-5.5 System Card<\/strong>\uff08OpenAI News\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>OpenAI \u53d1\u5e03 GPT-5.5 System Card\uff0c\u8be6\u7ec6\u62ab\u9732\u4e86\u8be5\u6a21\u578b\u7684\u5b89\u5168\u8bc4\u4f30\u4e0e\u90e8\u7f72\u51c6\u5907\u60c5\u51b5\u3002GPT-5.5 \u4e13\u4e3a\u590d\u6742\u73b0\u5b9e\u4e16\u754c\u4efb\u52a1\u8bbe\u8ba1\uff0c\u6db5\u76d6\u4ee3\u7801\u7f16\u5199\u3001\u5728\u7ebf\u7814\u7a76\u3001\u4fe1\u606f\u5206\u6790\u3001\u6587\u6863\u4e0e\u8868\u683c\u521b\u5efa\u53ca\u8de8\u5de5\u5177\u534f\u4f5c\u7b49\u573a\u666f\u3002\u76f8\u6bd4\u524d\u4ee3\u6a21\u578b\uff0c\u5b83\u80fd\u66f4\u5feb\u7406\u89e3\u4efb\u52a1\u3001\u51cf\u5c11\u7528\u6237\u6307\u5bfc\u3001\u66f4\u9ad8\u6548\u5730\u4f7f\u7528\u5de5\u5177\uff0c\u5e76\u5177\u5907\u81ea\u6211\u68c0\u67e5\u4e0e\u6301\u7eed\u8fed\u4ee3\u7684\u80fd\u529b\u3002OpenAI \u5728\u53d1\u5e03\u524d\u8fdb\u884c\u4e86\u5168\u9762\u7684\u9884\u90e8\u7f72\u5b89\u5168\u8bc4\u4f30\uff0c\u5305\u62ec\u9488\u5bf9\u9ad8\u7ea7\u7f51\u7edc\u5b89\u5168\u4e0e\u751f\u7269\u80fd\u529b\u7684\u4e13\u9879\u7ea2\u961f\u6d4b\u8bd5\uff0c\u5e76\u6536\u96c6\u4e86\u7ea6 200 \u5bb6\u65e9\u671f\u5408\u4f5c\u4f19\u4f34\u7684\u771f\u5b9e\u4f7f\u7528\u53cd\u9988\u3002\u8be5\u6a21\u578b\u642d\u8f7d\u4e86 OpenAI \u8fc4\u4eca\u4e3a\u6b62\u6700\u5b8c\u5584\u7684\u5b89\u5168\u9632\u62a4\u673a\u5236\uff0c\u65e8\u5728\u964d\u4f4e\u6ee5\u7528\u98ce\u9669\u7684\u540c\u65f6\u4fdd\u7559\u5408\u6cd5\u6709\u76ca\u7684\u9ad8\u7ea7\u80fd\u529b\u7528\u9014\u3002<\/p>\n

    English Summary:<\/strong> OpenAI released the GPT-5.5 System Card, detailing the model's safety evaluations and deployment readiness. GPT-5.5 is designed for complex real-world tasks including coding, online research, information analysis, document and spreadsheet creation, and cross-tool workflows. Compared to earlier models, it understands tasks faster, requires less guidance, uses tools more effectively, and can self-check and iterate until completion. OpenAI conducted comprehensive pre-deployment safety evaluations, including targeted red-teaming for advanced cybersecurity and biology capabilities, and gathered feedback from nearly 200 early-access partners. The model ships with OpenAI's strongest safeguards to date, balancing misuse prevention with preservation of legitimate advanced use cases.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  12. \n

    Reading today's open-closed performance gap<\/strong>\uff08Interconnects\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>Interconnects \u535a\u5ba2\u6587\u7ae0\u6df1\u5165\u5206\u6790\u4e86\u5f53\u524d\u5f00\u6e90\u4e0e\u95ed\u6e90\u5927\u6a21\u578b\u4e4b\u95f4\u7684\u6027\u80fd\u5dee\u8ddd\u8fd9\u4e00\u590d\u6742\u8bae\u9898\u3002\u4f5c\u8005\u6307\u51fa\uff0c\u867d\u7136\u5f00\u6e90\u6a21\u578b\u5728\u8ffd\u8d76\u95ed\u6e90\u6a21\u578b\uff0c\u4f46\u5c06\u8fd9\u4e00\u5dee\u8ddd\u7b80\u5316\u4e3a\u5355\u4e00\u6570\u5b57\u4f1a\u63a9\u76d6\u6a21\u578b\u80fd\u529b\u8986\u76d6\u8303\u56f4\u7684\u7ec6\u5fae\u5dee\u522b\u3002\u6587\u7ae0\u63a2\u8ba8\u4e86\u8bc4\u6d4b\u57fa\u51c6\u968f\u65f6\u95f4\u6f14\u53d8\u7684\u95ee\u9898\u3001\u6a21\u578b\u771f\u5b9e\u4e16\u754c\u8868\u73b0\u4e0e\u57fa\u51c6\u6392\u540d\u7684\u5173\u7cfb\uff0c\u4ee5\u53ca\u8bad\u7ec3\u65b9\u6cd5\u7684\u53d8\u5316\u5982\u4f55\u5f71\u54cd\u57fa\u51c6\u5206\u6570\u3002\u4f5c\u8005\u7279\u522b\u63d0\u5230\uff0c\u5f53\u524d\u8bc4\u6d4b\u57fa\u51c6\u5728\u8861\u91cf\u590d\u6742\u667a\u80fd\u4f53\u4efb\u52a1\u65b9\u9762\u5b58\u5728\u5c40\u9650\uff0c\u5982 Gemini 3 \u5728\u57fa\u51c6\u6d4b\u8bd5\u4e2d\u8868\u73b0\u51fa\u8272\u5374\u5728\u5b9e\u9645\u667a\u80fd\u4f53\u5e94\u7528\u573a\u666f\u4e2d relevance \u6709\u9650\u3002\u6587\u7ae0\u8fd8\u5206\u6790\u4e86\u5f00\u6e90\u5b9e\u9a8c\u5ba4\u9762\u4e34\u7684\u6570\u636e\u83b7\u53d6\u6311\u6218\uff0c\u4ee5\u53ca\u95ed\u6e90\u5b9e\u9a8c\u5ba4\u4e3a\u7ef4\u6301\u5546\u4e1a\u4f18\u52bf\u9700\u8981\u4e0d\u65ad\u91cd\u65b0\u5b9a\u4e49"\u524d\u6cbf"\u4efb\u52a1\u9886\u57df\u7684\u7ecf\u6d4e\u538b\u529b\u3002<\/p>\n

    English Summary:<\/strong> An Interconnects blog post provides an in-depth analysis of the complex issue of performance gaps between open and closed-source large language models. The author notes that while open models are catching up to closed ones, reducing this gap to a single number obscures nuanced differences in capability coverage. The article examines how benchmarks evolve over time, the relationship between real-world model performance and benchmark rankings, and how changes in training methodologies affect benchmark scores. The author specifically highlights limitations in current benchmarks for measuring complex agentic tasks, citing Gemini 3's strong benchmark performance but limited relevance in practical agent deployment scenarios. The piece also analyzes data acquisition challenges facing open labs and the economic pressure on closed labs to constantly redefine "frontier" task domains to maintain commercial advantages.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  13. \n

    Building an emoji list generator with the GitHub Copilot CLI<\/strong>\uff08GitHub AI\/ML\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>GitHub \u535a\u5ba2\u5206\u4eab\u4e86\u5728 Rubber Duck Thursday \u76f4\u64ad\u6d3b\u52a8\u4e2d\u4f7f\u7528 GitHub Copilot CLI \u6784\u5efa Emoji \u5217\u8868\u751f\u6210\u5668\u7684\u5b9e\u8df5\u6848\u4f8b\u3002\u8be5\u9879\u76ee\u662f\u4e00\u4e2a\u7ec8\u7aef\u5e94\u7528\uff0c\u7528\u6237\u53ef\u7c98\u8d34\u6216\u8f93\u5165\u5217\u8868\u9879\uff0c\u901a\u8fc7 Ctrl+S \u89e6\u53d1 AI \u81ea\u52a8\u751f\u6210\u76f8\u5173\u8868\u60c5\u7b26\u53f7\uff0c\u5e76\u5c06\u7ed3\u679c\u590d\u5236\u5230\u526a\u8d34\u677f\u3002\u9879\u76ee\u6280\u672f\u6808\u5305\u62ec @opentui\/core \u7528\u4e8e\u7ec8\u7aef UI\u3001@github\/copilot-sdk \u63d0\u4f9b AI \u80fd\u529b\u3001clipboardy \u5904\u7406\u526a\u8d34\u677f\u64cd\u4f5c\u3002\u5f00\u53d1\u8fc7\u7a0b\u5c55\u793a\u4e86 Copilot CLI \u7684 Plan \u6a21\u5f0f\uff08\u4f7f\u7528 Claude Sonnet 4.6 \u89c4\u5212\uff09\u4e0e Autopilot \u6a21\u5f0f\uff08\u4f7f\u7528 Claude Opus 4.7 \u5b9e\u73b0\uff09\u7684\u591a\u6a21\u578b\u534f\u4f5c\u6d41\u7a0b\uff0c\u540c\u65f6\u7ed3\u5408\u4e86 GitHub MCP Server \u548c allow-all \u5de5\u5177\u6807\u5fd7\u7b49\u7279\u6027\u3002\u8be5\u9879\u76ee\u5df2\u5f00\u6e90\uff0c\u5c55\u793a\u4e86 AI \u8f85\u52a9\u5f00\u53d1\u5c0f\u578b\u5b9e\u7528\u5de5\u5177\u7684\u9ad8\u6548\u6d41\u7a0b\u3002<\/p>\n

    English Summary:<\/strong> GitHub's blog shares a case study of building an Emoji List Generator using the GitHub Copilot CLI during their Rubber Duck Thursday livestream. The project is a terminal application where users can paste or type list items, trigger AI-powered emoji generation with Ctrl+S, and copy the results to clipboard. The tech stack includes @opentui\/core for terminal UI, @github\/copilot-sdk for AI capabilities, and clipboardy for clipboard operations. The development process showcased Copilot CLI's Plan mode (using Claude Sonnet 4.6 for planning) and Autopilot mode (using Claude Opus 4.7 for implementation) in a multi-model workflow, combined with features like the GitHub MCP Server and allow-all tools flag. The project is open-sourced, demonstrating an efficient AI-assisted development workflow for small utility tools.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  14. \n

    Build a personal organization command center with GitHub Copilot CLI<\/strong>\uff08GitHub AI\/ML\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>GitHub \u5de5\u7a0b\u5e08 Brittany Ellich \u5206\u4eab\u4e86\u4f7f\u7528 GitHub Copilot CLI \u6784\u5efa\u4e2a\u4eba\u7ec4\u7ec7\u6307\u6325\u4e2d\u5fc3\u7684\u7ecf\u9a8c\u3002\u8be5\u5de5\u5177\u65e8\u5728\u89e3\u51b3\u6570\u5b57\u4fe1\u606f\u5206\u6563\u95ee\u9898\uff0c\u5c06\u5206\u6563\u5728\u5341\u4f59\u4e2a\u5e94\u7528\u4e2d\u7684\u4fe1\u606f\u6574\u5408\u5230\u4e00\u4e2a\u7edf\u4e00\u7684\u4e2d\u592e\u7a7a\u95f4\u3002Ellich \u91c7\u7528"\u5148\u89c4\u5212\u540e\u5b9e\u73b0"\u7684\u5de5\u4f5c\u6d41\u7a0b\uff0c\u5229\u7528 AI \u8fdb\u884c\u89c4\u5212\u3001Copilot \u8fdb\u884c\u5b9e\u73b0\uff0c\u4ec5\u7528\u4e00\u5929\u65f6\u95f4\u5c31\u5b8c\u6210\u4e86 v1 \u7248\u672c\u3002\u5728\u89c4\u5212\u9636\u6bb5\uff0cCopilot \u901a\u8fc7\u63d0\u95ee\u6f84\u6e05\u9700\u6c42\uff0c\u751f\u6210 plan.md \u6587\u6863\u4ee5\u51cf\u5c11\u5b9e\u73b0\u9636\u6bb5\u7684\u731c\u6d4b\u3002\u5979\u504f\u597d\u7684\u5de5\u5177\u6808\u5305\u62ec VS Code \u7684 Agent \u6a21\u5f0f\u8fdb\u884c\u540c\u6b65\u5f00\u53d1\uff08\u540c\u65f6\u8fd0\u884c\u6700\u591a\u4e24\u4e2a\u975e\u7ade\u4e89\u6027\u7684 Agent \u5de5\u4f5c\u6d41\uff09\uff0c\u4ee5\u53ca Copilot Cloud Agent \u5904\u7406\u5f02\u6b65\u4efb\u52a1\u3002\u8be5\u9879\u76ee\u57fa\u4e8e Electron \u6784\u5efa\uff0c\u867d\u7136\u5927\u90e8\u5206\u4ee3\u7801\u7531 Agent \u751f\u6210\uff0c\u4f46\u5979\u624b\u52a8\u7b80\u5316\u4e86\u4ee3\u7801\u5e93\u4ee5\u63d0\u9ad8\u53ef\u7ef4\u62a4\u6027\u3002<\/p>\n

    English Summary:<\/strong> GitHub engineer Brittany Ellich shares her experience building a personal organization command center using the GitHub Copilot CLI. The tool addresses digital fragmentation by unifying information scattered across a dozen apps into one central space. Ellich employed a "plan-then-implement" workflow, using AI for planning and Copilot for implementation, completing v1 in just one day. During planning, Copilot interviewed her with clarifying questions to generate a plan.md document, reducing guesswork during implementation. Her preferred tool stack includes VS Code's Agent mode for synchronous development (running up to two non-competing agent workflows) and Copilot Cloud Agent for asynchronous tasks. Built on Electron, most code was generated by the Agent, though she manually simplified the codebase for better maintainability.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n

  15. \n

    Ollama is now powered by MLX on Apple Silicon in preview<\/strong>\uff08Ollama Blog\uff09<\/p>\n

    \u4e2d\u6587\u6458\u8981\uff1a<\/strong>Ollama \u53d1\u5e03\u9884\u89c8\u7248\u66f4\u65b0\uff0c\u5728 Apple Silicon \u4e0a\u96c6\u6210 MLX \u6846\u67b6\u4ee5\u5b9e\u73b0\u663e\u8457\u6027\u80fd\u63d0\u5347\u3002\u8be5\u66f4\u65b0\u5229\u7528 Apple \u7edf\u4e00\u5185\u5b58\u67b6\u6784\uff0c\u5728 M5\u3001M5 Pro \u548c M5 Max \u82af\u7247\u4e0a\u501f\u52a9\u65b0\u7684 GPU Neural Accelerator \u52a0\u901f\u9996 token \u751f\u6210\u65f6\u95f4\u548c\u751f\u6210\u901f\u5ea6\u3002\u6d4b\u8bd5\u663e\u793a\uff0c\u4f7f\u7528 Alibaba Qwen3.5-35B-A3B \u6a21\u578b\u7684 NVFP4 \u91cf\u5316\u7248\u672c\uff0c\u6027\u80fd\u76f8\u6bd4\u4e4b\u524d\u7684 Q4_K_M \u91cf\u5316\u5b9e\u73b0\u5927\u5e45\u63d0\u5347\u3002Ollama 0.19 \u7248\u672c\u8fd8\u5c06\u652f\u6301\u66f4\u9ad8\u7684\u6027\u80fd\u8868\u73b0\u3002\u6b64\u5916\uff0c\u8be5\u7248\u672c\u5f15\u5165 NVIDIA NVFP4 \u683c\u5f0f\u652f\u6301\uff0c\u5728\u4fdd\u6301\u6a21\u578b\u7cbe\u5ea6\u7684\u540c\u65f6\u964d\u4f4e\u5185\u5b58\u5e26\u5bbd\u548c\u5b58\u50a8\u9700\u6c42\uff0c\u4f7f\u672c\u5730\u63a8\u7406\u7ed3\u679c\u4e0e\u751f\u4ea7\u73af\u5883\u4fdd\u6301\u4e00\u81f4\u3002\u7f13\u5b58\u7cfb\u7edf\u4e5f\u5f97\u5230\u5347\u7ea7\uff0c\u5305\u62ec\u8de8\u4f1a\u8bdd\u7f13\u5b58\u590d\u7528\u3001\u667a\u80fd\u68c0\u67e5\u70b9\u5b58\u50a8\u548c\u66f4\u667a\u80fd\u7684\u7f13\u5b58\u6dd8\u6c70\u7b56\u7565\uff0c\u7279\u522b\u4f18\u5316\u4e86\u7f16\u7801\u548c\u667a\u80fd\u4f53\u4efb\u52a1\u7684\u6548\u7387\u3002<\/p>\n

    English Summary:<\/strong> Ollama released a preview update integrating the MLX framework on Apple Silicon for significant performance improvements. The update leverages Apple's unified memory architecture, utilizing new GPU Neural Accelerators on M5, M5 Pro, and M5 Max chips to accelerate both time-to-first-token and generation speed. Testing with Alibaba's Qwen3.5-35B-A3B model in NVFP4 quantization showed substantial performance gains over the previous Q4_K_M implementation, with Ollama 0.19 promising even higher performance. The release also introduces NVIDIA NVFP4 format support, maintaining model accuracy while reducing memory bandwidth and storage requirements for inference workloads, ensuring parity with production environments. The caching system has been upgraded with cross-session cache reuse, intelligent checkpoint storage, and smarter eviction policies, specifically optimizing efficiency for coding and agentic tasks.<\/p>\n

    \u539f\u6587\u94fe\u63a5<\/a><\/p>\n<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"

    \u65e5\u671f\uff1a2026-04-27 \u672c\u671f\u805a\u7126\uff1a\u91cd\u70b9\u5173\u6ce8\u6a21\u578b\u53d1\u5e03\u4e0e release notes\u3001\u5b98\u65b9 engineeri […]<\/p>\n","protected":false},"author":0,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[],"class_list":["post-357","post","type-post","status-publish","format-standard","hentry","category-ai-daily"],"_links":{"self":[{"href":"http:\/\/www.faiyi.com\/index.php?rest_route=\/wp\/v2\/posts\/357","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/www.faiyi.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.faiyi.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"http:\/\/www.faiyi.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=357"}],"version-history":[{"count":3,"href":"http:\/\/www.faiyi.com\/index.php?rest_route=\/wp\/v2\/posts\/357\/revisions"}],"predecessor-version":[{"id":360,"href":"http:\/\/www.faiyi.com\/index.php?rest_route=\/wp\/v2\/posts\/357\/revisions\/360"}],"wp:attachment":[{"href":"http:\/\/www.faiyi.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=357"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.faiyi.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=357"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.faiyi.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=357"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}