全部未入选
即梦官网功能监测 · 10-28 23:00 UTC
这是一条关于即梦官网页面内容变化的监测线索，所展示的主要是文生图、文/图生视频等现有功能介绍，尚未核实是否为新功能或新发布。
AI 判断：来源明确标注待核实，内容仅为官网现有功能文案，缺乏可演示的新能力、具体教程和发布时间，实用性与时效性都较弱。
展开原始摘录
这是官网内容变化线索，观察时间不是产品发布日期；尚未核实是否为新功能，不应直接称为今日发布。
即刻造梦 前往即梦 水母在海底遨游 即梦成片 此视频内容由 即梦AI 生成 支持文/图 生视频 灵感即刻成片 AI视频生成，输入简单的文案或图片，即可快速生成优质视频片段 流畅运镜，生动自然 视频动效效果连贯性强、流畅自然，可轻松操控运镜，调节速度变化，即梦AI为视频智能创作增添无限可能 立即创作 01 从首帧到尾帧， 精准掌控 创新的首帧图片和尾帧图片输入方式，增强视频生成的可控性，轻松打造高品质素材，提升你的视频内容创作效率 立即创作 02 中文创作，得心应手 即梦AI支持根据中文提示词进行创作，拥有更好的语义理解能力，准确把握你的需求，将抽象的思路转化为视觉作品 立即创作 03 支持文/图生图片 AI绘画 梦境成真 即梦AI绘画，仅需简单提示词，即可生成精彩的图片。 你还可以对现有图片进行创意改造，自定义保留人物或主体的形象特征，实现背景替换、风格联想、画风保持、姿势保持等 操作，满足各种场景的创作需求 3D游戏人物风格，戴夸张耳饰的民族风少女 即梦成片 插画风格，太空飞行器在粉色星球低空飞行 即梦成片 未来风，女性机械人形AI 即梦成片 超写实插画，夜晚月光下的花园 即梦成片 文艺复兴风格，身着贵族丝绸服饰猫咪 即梦成片 电影质感，盔甲勇士在火光下奔赴战场 即梦成片 即梦成片 智能画布 多图AI融合 即梦AI一站式智能画布，集成AI拼图生成能力，并提供局部重绘、一键扩图、图像消除和抠图等多功能强大操作。你可以在同一画布上实现多元素的无缝拼接，确保AI绘画的创作风格统一和谐。 立即创作 多图层编辑 精细化控制 连续包年 限时4.5折 连续包月 4.8折 连续包季 限时4.3折 单月购买 基础会员 ¥ 393 每年 首年6折¥393 · 下年续费金额¥659 · 可随时取消 725 积分每月 换算¥10=221积分 立即体验 ✓ Seedance 2.0 VIP 专属通道 ✓ 标准生成通道 ✓ 生成作品去除品牌水印 ✓ 查看全部权益 标准会员 ¥ 1099 每年 首年5.8折¥1099 · 下年续费金额¥1899 · 可随时取消 2210 积分每月 换算¥10=241积分 立即体验 ✓ Seedance 2.0 VIP 专属通道 ✓ 标准生成通道 ✓ 生成作品去除品牌水印 ✓ 查看全部权益 Seedance 2.5 720P限时低至￥0.43/秒 高级会员 限时5折 ¥ 5199 每年 首年5折¥5199 · 下年续费金额¥10398 · 可随时取消 6.2K 12.3K 18.5K 27.7K 12320 积分每月 换算¥10=284积分 立即体验 专享权益 备战双11 首3月16020积分/月 首3月加赠后16020积分/月！SD 2.5 720P带参考视频消耗6.2折 ￥0.43/秒，CLI调用折上9折 ￥0.38/秒 (10月13日-10月29日) ✓ Seedance 2.0 VIP 专属通道 ✓ 图片4.0 4K 限时免费 ✓ 快速生成通道 ✓ 生成作品去除品牌水印 ✓ 查看全部权益 Seedance 2.5 720P限时低至￥0.19/秒 超级会员 限时4.5折 ¥ 19656 每年 首年4.5折¥19656 · 下年续费金额¥43680 · 可随时取消 54600 积分每月 换算¥10=333积分 立即体验 专享权益 备战双11 首3月96060积分/月 1)多得2个月会员；2)首3月得96060积分/月；3)SD2.5 720P带参考视频消耗5折，CLI调用折上9折 ￥0.18/秒 (10月13日-10月29日) ✓ Seedance 2.0 VIP 专属通道 ✓ 图片4.0 4K 限时免费 ✓ 快速生成通道 ✓ 生成作品去除品牌水印 ✓ API服务购买 ✓ 查看全部权益 连续包月 基础会员 ¥ 33 每月 首月4.8折¥33 · 下月续费金额¥69 · 可随时取消 725 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月725积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 标准会员 ¥ 96 每月 首月4.9折¥96 · 下月续费金额¥199 · 可随时取消 2210 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月2210积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 高级会员 6.2K ¥ 259 每月 首月5.2折¥259 · 下月续费金额¥499 · 可随时取消 6160 积分每月 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月6160积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 12.3K ¥ 519 每月 首月5.3折¥519 · 下月续费金额¥998 · 可随时取消 12320 积分每月 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月12320积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 18.5K ¥ 779 每月 首月5.3折¥779 · 下月续费金额¥1498 · 可随时取消 18480 积分每月 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月18480积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 27.7K ¥ 1168 每月 首月5.3折¥1168 · 下月续费金额¥2246 · 可随时取消 27720 积分每月 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月27720积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 超级会员 ¥ 2235 每月 首月5.2折¥2235 · 下月续费金额¥4299 · 可随时取消 54600 积分每月 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 API服务购买 购买API套餐且持续为会员状态时，可享受API服务，使用额度需单独购买，与会员积分不通用。 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月54600积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 连续包季 基础会员 ¥ 188 每季 每月¥62.66 · 自动续订，可随时取消 725 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月725积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 标准会员 ¥ 568 每季 每月¥189.33 · 自动续订，可随时取消 2210 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月2210积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 高级会员 6.2K Seedance 2.5 720P限时低至￥1.21/秒 ¥ 1399 每季 每月¥466.33 · 自动续订，可随时取消 6160 积分每月 专享权益 Seedance 2.5 720P带参考视频消耗6.2折，低至￥1.21/秒 (10月13日-10月29日) Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月6160积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 12.3K Seedance 2.5 720P限时低至￥0.45/秒 限时4.5折 ¥ 1259 每季 首季4.5折¥1259 · 下季续费金额¥2798 · 可随时取消 12320 积分每月 专享权益 首月加赠2460积分，共得14780积分！再享Seedance 2.5 720P带参考视频消耗6.2折，低至￥0.45/秒 (10月13日-10月29日) 首月14780积分 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外赠送积分每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月12320积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 18.5K Seedance 2.5 720P限时低至￥0.4/秒 限时4.5折 ¥ 1890 每季 首季4.6折¥1890 · 下季续费金额¥4199 · 可随时取消 18480 积分每月 专享权益 首月加赠6470积分，共得24950积分！再享Seedance 2.5 720P带参考视频消耗6.2折，低至￥0.4/秒 (10月13日-10月29日) 首月24950积分 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外赠送积分每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月18480积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 27.7K Seedance 2.5 720P限时低至￥0.39/秒 限时4.5折 ¥ 2834 每季 首季4.5折¥2834 · 下季续费金额¥6298 · 可随时取消 27720 积分每月 专享权益 首月加赠11080积分，共得38800积分！再享Seedance 2.5 720P带参考视频消耗6.2折，低至￥0.39/秒 (10月13日-10月29日) 首月38800积分 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外赠送积分每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月27720积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 超级会员 Seedance 2.5 720P限时低至￥0.27/秒 限时4.3折 ¥ 4999 每季 首季4.3折¥4999 · 下季续费金额¥11699 · 可随时取消 54600 积分每月 专享权益 首月加赠26000积分，共得80600积分！再享Seedance 2.5 720P带参考视频消耗5折，低至￥0.27/秒 (10月13日-10月29日) 首月80600积分 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外赠送积分每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 API服务购买 购买API套餐且持续为会员状态时，可享受API服务，使用额度需单独购买，与会员积分不通用。 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月54600积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 单月购买 基础会员 ¥ 79 每月 一次性支付¥79，有效期1个月，不自动续费 725 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月725积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 标准会员 ¥ 239 每月 一次性支付¥239，有效期1个月，不自动续费 2210 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月2210积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 高级会员 ¥ 649 每月 一次性支付¥649，有效期1个月，不自动续费 6160 积分每月 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月6160积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 连续包年 基础会员 ¥ 393 每年 首年6折¥393 · 下年续费金额¥659 · 可随时取消 725 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月725积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 标准会员 ¥ 1099 每年 首年5.8折¥1099 · 下年续费金额¥1899 · 可随时取消 2210 积分每月 Seedance 2.0 VIP 专属通道 标准生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月2210积分 seedance2.0 VIP模型：✓ 视频生成专享加速：标准生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 智能超清：4K 去除品牌水印：✓ 生成加速：✓ 高级会员 6.2K Seedance 2.5 720P限时低至￥0.47/秒 限时5折 ¥ 2599 每年 首年5折¥2599 · 下年续费金额¥5199 · 可随时取消 6160 积分每月 专享权益 首3月加赠后7410积分/月！SD 2.5 720P带参考视频消耗6.2折 ￥0.47/秒，CLI调用折上9折 ￥0.41/秒 (10月13日-10月29日) 首3月7410积分/月 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外积分连续加赠3个月，每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月6160积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 12.3K Seedance 2.5 720P限时低至￥0.43/秒 限时5折 ¥ 5199 每年 首年5折¥5199 · 下年续费金额¥10398 · 可随时取消 12320 积分每月 专享权益 首3月加赠后16020积分/月！SD 2.5 720P带参考视频消耗6.2折 ￥0.43/秒，CLI调用折上9折 ￥0.38/秒 (10月13日-10月29日) 首3月16020积分/月 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外积分连续加赠3个月，每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月12320积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 18.5K Seedance 2.5 720P限时低至￥0.36/秒 限时4.5折 ¥ 7019 每年 首年4.5折¥7019 · 下年续费金额¥15598 · 可随时取消 18480 积分每月 专享权益 首3月加赠后25880积分/月！SD 2.5 720P带参考视频消耗6.2折 ￥0.36/秒，CLI调用折上9折 ￥0.32/秒 (10月13日-10月29日) 首3月25880积分/月 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外积分连续加赠3个月，每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月18480积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 高级会员 27.7K Seedance 2.5 720P限时低至￥0.34/秒 限时4.5折 ¥ 10529 每年 首年4.5折¥10529 · 下年续费金额¥23398 · 可随时取消 27720 积分每月 专享权益 首3月加赠后41620积分/月！SD 2.5 720P带参考视频消耗6.2折 ￥0.34/秒，CLI调用折上9折 ￥0.30/秒 (10月13日-10月29日) 首3月41620积分/月 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外积分连续加赠3个月，每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月27720积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 超级会员 Seedance 2.5 720P限时低至￥0.19/秒 限时4.5折 ¥ 19656 每年 首年4.5折¥19656 · 下年续费金额¥43680 · 可随时取消 54600 积分每月 专享权益 1)多得2个月会员；2)首3月得96060积分/月；3)SD2.5 720P带参考视频消耗5折，CLI调用折上9折 ￥0.18/秒 (10月13日-10月29日) 首3月96060积分/月 1. 加赠福利为大型促销活动，暂不支持退款。法律法规另有规定除外。 2. 额外积分连续加赠3个月，每月跟随订阅积分发放，非一次性发放，有效期为29天。3. 特别注意：每个账户仅能享受1次加赠福利，多次购买不叠加，退款后重新购买不重新发放。一旦会员失效，对应赠送的福利即时终止。4. 单秒价格计算方式为：模型1秒消耗积分数*（活动价/订阅周期/获得积分数），所得结果四舍五入保留两位小数 Seedance 2.0 VIP 专属通道 图片4.0 4K 限时免费 持续为高级会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 快速生成通道 生成作品去除品牌水印 API服务购买 购买API套餐且持续为会员状态时，可享受API服务，使用额度需单独购买，与会员积分不通用。 平台免费积分：每日登录赠送积分 充值购买积分：✓ 订阅会员积分：每月54600积分 seedance2.0 VIP模型：✓ 视频生成专享加速：快速生成通道 视频对口型：✓ 视频高清：✓ 视频补帧：✓ 图片 4.0 限时免费：4K 持续为会员状态时，可享受图片 4.0 4k生成免费至2026.11.06 智能超清：4K/8K 去除品牌水印：✓ 生成加速：✓ 创意涌动 灵感绽放 在即梦AI的创意社区，与其他用户共同探索无限的影像灵感。让创意在相互碰撞中激发，一起创造精彩 获取创作灵感 🍰Berry 16 创作同款 盒饭饭喜欢吃丸子 26 创作同款 小树树爱世界！ 56 创作同款 晓梦 35 创作同款 公主 34 创作同款 塔尼雅 10 创作同款 不吃草的 花牛 25 创作同款 水水 19 创作同款 晗槑槑 17 创作同款 🍰Berry 10 创作同款 檬檬 5 创作同
爱范儿 · 10-28 22:31 UTC
标题称 Gemini 4 Carbon 已正式发布，并宣称哈基米重回世界第一。
AI 判断：标题初筛（未依据正文）：仅凭标题只能看到模型发布和排名宣称，缺少普通用户可用功能、价格与验证信息，暂按技术发布处理。
展开原始摘录
哈基米终于扬眉吐气了
#欢迎关注爱范儿官方微信公众号：爱范儿（微信号：ifanr），更多精彩内容第一时间为您奉上。
Google Expands AI Mode Ads to 10 New Countries Including Canada and India
Future Tools · 10-28 22:00 UTC
谷歌宣布把 AI 模式里的广告投放到澳大利亚、加拿大、印度、肯尼亚等 10 个新市场，并称这些广告的点击率与自然搜索结果相当。
AI 判断：这是 AI 搜索商业化的地区性进展，与中国用户日常使用无关且难以亲自演示，实用性和可看性都有限，只能作为趋势话题带一句。
展开原始摘录
Company News · 1h ago Google Expands AI Mode Ads to 10 New Countries Including Canada and India Google is expanding ads in AI Mode to 10 new countries: Australia, Canada, India, Kenya, Malaysia, New Zealand, Nigeria, Pakistan, the Philippines, and Singapore. Announced by VP Shashi Thakur, the rollout follows research showing users find the ads highly relevant, with click-through rates comparable to organic results. AI Mode queries average three times longer than traditional searches. Google also plans further expansion to Brazil, Mexico, and the UK in Q4 2026, and is launching multiturn ads in the U.S. for follow-up queries.
Google Official Blog · 10-28 21:30 UTC · 同链接合并评分
谷歌宣布把 AI 模式里的广告投放到澳大利亚、加拿大、印度、肯尼亚等 10 个新市场，并称这些广告的点击率与自然搜索结果相当。
AI 判断：这是 AI 搜索商业化的地区性进展，与中国用户日常使用无关且难以亲自演示，实用性和可看性都有限，只能作为趋势话题带一句。
展开原始摘录
Google is expanding AI Mode ads to 10 new countries, including Canada and India. See how these relevant sponsored results improve your search experience.
Leak Exposes 41,000 Homeland Security Workers–Including Undercover ICE Agents
Aboutplants · 10-28 21:03 UTC
Wired 报道称，一次大规模数据泄露暴露了约4.1万名美国国土安全部员工的信息，其中包含 ICE 卧底探员。
AI 判断：这是重要的隐私与安全新闻，但与面向普通人的 AI 工具或应用无关，也缺乏可演示的实用价值，因此评分较低。
展开原始摘录
--- Top Comments ---
[sailfast]: [delayed]
[paaloeye]: So cool, terrorist Karen can now easily email/LinkedIn/whatever escalate to a skip-level manager if an officer is misbehaving.
[fastily]: The primary source for the article: https://ddosecrets.org/article/icepick
[JumpCrisscross]: Can you search for agents near me?
[resters]: At first I hoped that fears about ICE abuses were overblown, but then I saw video after video of masked, badge-free agents manhandling and throwing grandmas and elderly/weak people to the ground and generally acting like thugs, stalking immigration courts for people trying to comply, doing nazi salutes, etc. It quickly became clear that ICE is largely made up of KKK or proud boy thugs who see it as a chance to brutalize a vulnerable population that they despise. Now that this data has b...
Alphabet Q3 财报：年化收入达 5000 亿美元，Google Cloud 增长 103%
X：Sundar Pichai (@sundarpichai) · 10-28 20:05 UTC
Alphabet 公布第三季度财报，称年化收入达到 5000 亿美元，Google Cloud 增长 103%。
AI 判断：标题初筛（未依据正文）：仅凭标题看，这是财报业绩数据，缺少普通人可直接使用或演示的 AI 功能，属于应降权的融资业绩类新闻。
展开原始摘录
Sundar Pichai 发布 Alphabet Q3 业绩：年化收入 run rate 达 5000 亿美元，总收入同比增长 25%。Google Cloud 收入增长 103%，1P 模型 API 每分钟处理近 32B tokens（环比增 40%）；Search 查询同比增速为五年来最高，并提到 Argon、Carbon 领跑前沿及最新世界模型 Genie 4。
ChatGPT's "dot" AI Can Now Text Users Proactively via iMessage or RCS for Pro Members
Future Tools · 10-28 18:00 UTC
OpenAI 为 ChatGPT 的「dot」AI 推出主动发消息预览功能，美国 Pro 用户可让它盯任务并通过 iMessage 或 RCS 主动推送文字、图片和语音更新。
AI 判断：功能方向有意思（AI 从被动问答转向主动推送），但仅限美国 Pro 用户预览、国内无法亲自演示，实用性与画面感受限。
展开原始摘录
Tools & Products · 5h ago ChatGPT's "dot" AI Can Now Text Users Proactively via iMessage or RCS for Pro Members OpenAI has launched a proactive texting feature for ChatGPT's "dot" AI, now available as a preview for Pro users in the US. Users can assign tasks or ask dot to monitor something, and it will send updates via iMessage or RCS without being prompted. Texts, photos, and voice notes are supported. The feature is private — only the user can text their dot and vice versa — and OpenAI does not gain access to other text conversations. Set it up via dot's profile under 'Texting'.
Gemini App Launches Immersive Explorations for Interactive Visual Learning
Future Tools · 10-28 18:00 UTC
Gemini 应用在“学生中心”里新增沉浸式探索功能，用户可点标签逐层缩放，从行星尺度一路看到分子结构，首批主题包括丝绸之路、喷气发动机和闪电原理。
AI 判断：这是一项普通人可上手的可视化学习新功能，画面感强、便于录屏演示，但目前仅支持英语、印尼语、印地语、巴西葡萄牙语等，中文可用性尚不确定。
展开原始摘录
Tools & Products · 5h ago Gemini App Launches Immersive Explorations for Interactive Visual Learning Google's Gemini app has launched immersive explorations, an interactive visual learning feature that lets users explore topics layer-by-layer, from microscopic systems to deep space. Accessible via the Gemini student hub under "Try immersive explorations," topics include the Silk Road, jet engines, and lightning physics. Users can tap labels to zoom from planetary scales down to molecules. Audio narration is coming soon. The feature is available in English, Indonesian, Hindi, Brazilian Portuguese, French, and simplified Chinese, with more languages planned.
Anthropic Launches Claude for Legal Artifacts: Dashboards, Contract Review & 60+ Connectors
Future Tools · 10-28 18:00 UTC
Anthropic 扩展 Claude for Legal，新增可复用的交互式工件，覆盖合同审查、监管追踪与案件管理，并接入 60 多个法律数据连接器。
AI 判断：这是面向法律团队的专业功能更新，普通用户相关性低，也缺少可亲自演示的消费级使用场景。
展开原始摘录
Tools & Products · 5h ago Anthropic Launches Claude for Legal Artifacts: Dashboards, Contract Review & 60+ Connectors Anthropic has expanded Claude for Legal with a gallery of ready-to-use artifacts — interactive tools like dashboards and trackers — that legal teams can adapt and share across their organizations. Launched on October 28, 2026, the update includes artifacts for contract review, regulatory tracking, cite checking, and matter management. Since Claude for Legal debuted in May, Anthropic has added over 60 legal connectors, including Bloomberg Law, Filevine, and Docusign. Partner-built artifacts from Goodwin, Harvey, Foley & Lardner, and PwC are also available in the gallery.
Google Official Blog · 10-28 18:00 UTC · 同链接合并评分
Gemini 应用在“学生中心”里新增沉浸式探索功能，用户可点标签逐层缩放，从行星尺度一路看到分子结构，首批主题包括丝绸之路、喷气发动机和闪电原理。
AI 判断：这是一项普通人可上手的可视化学习新功能，画面感强、便于录屏演示，但目前仅支持英语、印尼语、印地语、巴西葡萄牙语等，中文可用性尚不确定。
展开原始摘录
Today, we are rolling out immersive explorations in the new Gemini student hub.
Why is this fingerprinting script creating bioweapons?
hazebooth · 10-28 17:13 UTC
文章探讨一个指纹脚本为何会触发“制造生物武器”相关判定，偏向安全技术排查。
AI 判断：属于安全/技术类排查，普通用户既难理解也难直接使用。
展开原始摘录
--- Top Comments ---
[jacquesm]: Nice sleuthing! We do a similar thing to mark documents that are still under embargo and one day I forgot to remove the watermark. The customer complained they had not received the document (a day or two after I sent it), and much searching was done. Finally we found a line in an outlook log on their end that said that the watermark was the reason the file had been put in indefinite quarantine until someone would take a look at it. That's when I realized my error, and of course it was so...
[throwaway89201]: Nice finds! This HN profile I looked at today by chance, also contains some anti-LLM incantations, albeit much less obfuscated. I wasn't aware Anthropic had such a magic string: https://news.ycombinator.com/user?id=jeroenhd
[ghm2199]: Whoa! This is insane. This is an insane arms race, fractal like craziness. Dynamic JS obfuscator libraries that are like honey pots: Lure LLM bots and determine if your obfuscation can make then drop the crawl attempt. So are now at RLing our obfuscators? But I have a more abstract question: Are LLMs mass "codemoding" scraping attempt in their context windows to make this work _and_ worth the effort? It seems like it is to prevent a different thing than what Anubis does(which is pre...
[arbol]: This stuff doesn't work. The agents read the AST and just ignore dead code. Even replacing all your function names at build time with decoys makes the agent see the juxtaposition. You need evil context as well as evil names, and then you need to run the evil code! Not an appropriate defense on client sites. To stop deobfuscation at scale, you're better off serving so many bundles that mass scraping can't scale, as they cannot reliably deobfuscate your code. This forces them int...
[Revanche1367]: Maybe it's a Russian bioweapons company masquerading as an Internet traffic control company pretending to be in bioweapons manufacturing so any snoopers think it's just anti-LLM obfuscation.
Anthropic IPO 文件披露现金及短期投资达 131.7B 美元，超过 Amazon 和 Microsoft
X：Rohan Paul (@rohanpaul_ai) · 10-28 17:11 UTC
据标题称，Anthropic 的 IPO 文件显示其现金及短期投资规模达 1317 亿美元，高于亚马逊和微软。
AI 判断：标题初筛（未依据正文）：这是纯财务与融资层面的消息，普通人无法据此获得可用的 AI 功能或用法，且数字仅来自标题、未核实。
展开原始摘录
Anthropic 的 IPO 文件显示其现金及短期投资达 $131.7B，超过 Amazon 和 Microsoft，在美国非银行公司中仅次于 Berkshire Hathaway 和 Alphabet。
MacBook Pro with PROLED XDR Display and Touch
etothet · 10-28 17:10 UTC
苹果推出搭载 PROLED XDR 显示屏并支持触控的 MacBook Pro，但现有信息未显示其与 AI 功能直接相关。
AI 判断：这是消费硬件发布，缺少可核实的 AI 用法或普通人可演示的 AI 价值，与 AI 实用选题关联较弱。
展开原始摘录
https://www.apple.com/newsroom/2026/10/apple-unveils-groundb...
--- Top Comments ---
[CountSessine]: Touch-screen Macs is kind of a terrible development for anyone who loves the Mac. It's not just a harmless feature that you can ignore if you don't want it, because it's customer-facing and will now generate complaints and bugs from people who do use it. Those people will complain (rightly) that macOS UI widgets are too small to accurately manipulate with your fingers (they are indeed too small) and don't make sense with a touch screen (they don't make sense). That ...
[tristanj]: Everyone's focused on the touchscreen, but I'm seriously impressed how they managed to cut down the weight. * For the 14-inch: 3.5 lb (1.59 kg) -> 3.0 lb (1.36 kg) * For the 16-inch: 4.7 lb (2.13 kg) -> 4.0 lb (1.81 kg) About 15% lighter on both models. Switching from LCD to OLED saves weight because the screen does not need a backlight and has fewer layers. They also made the laptop thinner by nearly 1 mm. But that's not enough to push the weight down by 300 grams. They ...
[Toutouxc]: Oh wow, it’s like they’re trying less and less. When they introduced the Apple Pencil and people were like “haha it’s a stylus”, the thing was decidedly NOT a stylus (it actually couldn’t be used to control some parts of the UI). It was an actual digital pen, something that used to be specialized and expensive external hardware, and it worked beautifully with the iPad. It was literally “stylus but GOOD”. I saw it and thought “people will make memes about this, but Apple totally gets it”. When...
[joshstrange]: Does anyone use/like touch on their windows (linux?) laptops? I've never really been drawn to wanting to do that. My mouse is far more accurate and doesn't block my view. But maybe it's just because I don't have it and I would find tons of uses for it. Also, I don't love the idea of putting fingerprints all over my screen. I guess I can understand for testing iOS/iPad apps but past that I just don't see myself wanting to use that. Then again, it really ...
[Mongoose]: Fully willing to accept that I could be an outlier, but I have never understood how a market exists for laptops with touch screens.
X：Boris Cherny (@bcherny) · 10-28 17:09 UTC
据标题信息，犹他州 10 个州政府部门的 34 名开发者正在试点使用 Claude Code。
AI 判断：标题初筛（未依据正文）：这是开发者工具在政府部门的试点消息，普通用户缺少可上手复现的场景，属技术类选题。
展开原始摘录
Boris Cherny 介绍犹他州向 10 个州政府部门的 34 名开发者提供了 Claude Code。试点显示 80% 的开发者在 30 分钟内搭好可用环境，71% 的开发者在为期四周的试点结束前报告节省了 20 小时以上时间。该推文引用了 Anthropic 与 GSA 合作将 Claude 以 40% 折扣提供给联邦及各级政府的公告作为背景。
Google 发布 Gemini 4 Carbon：AA Index v5.0 登顶，网络安全基准领先
X：Sundar Pichai (@sundarpichai) · 10-28 17:01 UTC
标题称谷歌发布 Gemini 4 Carbon，并称其在 AA Index v5.0 排名登顶、在网络安全基准上领先。
AI 判断：标题初筛（未依据正文）：仅凭标题只有模型发布与跑分领先信息，缺少普通人可用的功能、价格与地区说明，偏技术且无法演示（判断基于标题，榜单口径未核实）。
展开原始摘录
Sundar Pichai 宣布 Google 推出 Gemini 4 Carbon，称其在更新后的 Artificial Analysis Intelligence Index v5.0 以 59 分排名第一。
Google DeepMind：Blog（RSS） · 10-28 17:00 UTC
Google DeepMind 宣布发布名为 Gemini 4 Carbon 的前沿模型。
AI 判断：标题初筛（未依据正文）：仅凭标题可知这是前沿模型发布，未提供普通人可用功能、价格或演示，偏技术发布。
展开原始摘录
Google DeepMind 发布 Gemini 4 Carbon，定位为最强前沿模型，主打网络安全与长程智能体能力，在 SEC-bench Pro 88.2%、ExploitGym 54.8%、CWE-bench 71.0% 取得领先，并以 59 分登顶 Artificial Analysis Intelligence Index v5.0。
Future Tools · 10-28 17:00 UTC · 同链接合并评分
Google 官方博客介绍新模型 Gemini 4 Carbon。
AI 判断：官方模型发布信息有限，缺少普通用户可用的功能、价格与地区细节，实际价值尚不确定。
展开原始摘录
Model Release · 6h ago Matt's Pick Google Launches Gemini 4 Carbon With Top Cybersecurity and Long-Horizon AI Capabilities Google has announced Gemini 4 Carbon, its strongest frontier model, featuring state-of-the-art cybersecurity and long-horizon agentic capabilities. Carbon ranks #1 on the Artificial Analysis Intelligence Index v5.0 with a score of 59, and leads key security benchmarks: SEC-bench Pro (88.2%), ExploitGym (54.8%), and CWE-bench (71.0%). It also achieves 0.0% prompt injection success rate and reduces high-severity misalignment alerts by 80%. Priced at $2 per million input tokens and $10 per million output tokens, Carbon is initially rolling out to select cyber defenders via Google's Fairwind Program.
Google Launches Connected Apps Platform for Gemini & Search, Adds Adobe, Peloton Partners
Future Tools · 10-28 17:00 UTC
Google 预览面向开发者的 Connected Apps 平台，让开发者通过 MCP 为搜索和 Gemini 应用统一构建并管理第三方应用集成。
AI 判断：这是面向开发者的集成平台，普通用户只能间接受益，缺乏可亲自演示的消费级用法。
展开原始摘录
Tools & Products · 6h ago Google Launches Connected Apps Platform for Gemini & Search, Adds Adobe, Peloton Partners Google has launched a preview of its Connected Apps Platform, giving developers a unified way to build and manage integrations for both Search and the Gemini app. Using cloud-hosted MCP servers or Android MCP via AppFunctions APIs, developers register and submit apps through the Google Connected Console. New partners including Adobe, Experian, Peloton, Localiza, and Calm are now live or coming soon. Over 50 Connected Apps are available across productivity, health, and fitness categories.
Google Official Blog · 10-28 17:00 UTC · 同链接合并评分
Google 官方博客介绍新模型 Gemini 4 Carbon。
AI 判断：官方模型发布信息有限，缺少普通用户可用的功能、价格与地区细节，实际价值尚不确定。
展开原始摘录
Faceted iridescent crystal surrounded by floating white number 4s.
Google Official Blog · 10-28 17:00 UTC · 同链接合并评分
Google 预览面向开发者的 Connected Apps 平台，让开发者通过 MCP 为搜索和 Gemini 应用统一构建并管理第三方应用集成。
AI 判断：这是面向开发者的集成平台，普通用户只能间接受益，缺乏可亲自演示的消费级用法。
展开原始摘录
A split-screen graphic showing a Google search interface connecting to Gemini to confirm a dinner reservation.
Microsoft 通过 Foundry 开放 Copilot harness，支持多模型 Agent 开发
Microsoft：Official Blog（RSS） · 10-28 16:51 UTC
微软在官方博客宣布，通过 Foundry 开放 Copilot harness，用于支持多模型 Agent 的开发。
AI 判断：标题初筛（未依据正文）：这是仅面向开发者的 Agent 开发框架消息，标题中没有普通人能直接使用的功能或演示（仅基于标题判断）。
展开原始摘录
Microsoft 宣布将 GitHub 团队打造的 Copilot harness 通过 Microsoft Foundry 向开发者和企业开放，该 harness 已驱动 Copilot Code、Copilot Studio、GitHub Copilot app 和 CLI 等产品。
GitHub Octoverse 2026
Simon Willison · 10-28 16:45 UTC
GitHub 发布 2026 年度 Octoverse 报告，汇总其平台生态的年度数据。
AI 判断：标题初筛（未依据正文）：这是面向开发者的平台年度统计报告，普通用户难以直接使用或演示，且仅有标题、细节无法核实，因此按技术类低分处理。
展开原始摘录
GitHub Octoverse 2026
I'm at GitHub Universe today (GitHub gave me a free ticket through their "GitHub Stars" program). EVP Jay Parikh just announced this year's Octoverse report, and highlighted these numbers in particular:
This is a pretty stark representation of the impact of agentic engineering on GitHub. In 2026, merged pull requests tripled in less than a year to 220 million per month. Commits more than tripled to 7.38 billion per month, with most of that increase in the second half of this year. New repository rate doubled this year, to 30.3 million per month.
The report includes new numbers by industry, and breaks out numbers for repositories that involve "interoperability or reverse engineering". The
methodology section
includes notes about how they did this:
Top countries by pushes to public repositories, adjusted by profile location
. Pushes from each country were summed across the Octoverse year. To adjust for biases caused by VPN usage, LLM classification was used to extract the ISO 3166-1 alpha-2 code from the profiles of GitHub users who contributed pushes during this period [...]
Industry
. Public repositories created 2022 to 2026 were sorted into NAICS sectors by a fine-tuned multilingual embedding model using name, description, topics, and README, from random samples per year. [...]
Interoperability
. Counts public repositories created each year whose name, description, topics, or README mention interoperability or reverse engineering, or related terms, such as clean room implementation, drop-in replacement, or protocol re-implementation.
Tags:
github
,
ai
,
generative-ai
,
llms
,
coding-agents
,
disclosures
,
agentic-engineering
GitHub Universe 2026 发布 GitHub Copilot app、Copilot Fusion 等智能体产品
GitHub Blog · 10-28 16:09 UTC
GitHub 在 Universe 2026 大会上宣布推出 GitHub Copilot 应用和 Copilot Fusion 等智能体产品。
AI 判断：标题初筛（未依据正文）：标题显示这是面向开发者的智能体产品发布，普通人的直接使用价值和可演示性有限，且仅凭标题无法判断具体功能与使用限制。
展开原始摘录
GitHub 在 GitHub Universe 2026 上提出把 GitHub 打造为完整软件交付系统的愿景，并发布多项产品：GitHub Copilot app 面向智能体交付。
GitHub Octoverse 2026：开发活动创新高，超 40% 的 PR 由智能体起草
GitHub Blog · 10-28 16:07 UTC
GitHub年度Octoverse报告显示，2026年开发活动创下新高，超过40%的拉取请求由智能体起草。
AI 判断：标题初筛（未依据正文）：这是开发者平台的年度数据，普通人难以直接使用或亲自演示，因此按技术类内容处理并给予较低评分。
展开原始摘录
GitHub 发布 2026 年度 Octoverse 报告：过去 12 个月开发者推送代码 179 亿次，超过 2024 和 2025 年之和，合并 PR 达 13 亿次。
slowin · 10-28 16:01 UTC
Meta censored the No Kings Instagram account during a nationwide campaign
AI 判断：前置规则过滤：HN 标题及已抓取内容未命中 AI 主题
展开原始摘录
--- Top Comments ---
[aleksandrm]: Not the first time, and not the last. They just did it again yesterday to Democratic Senate candidate Angie Nixon. https://newrepublic.com/post/217281/meta-suspend-democratic-...
[Ancalagon]: Getting the sense Meta's internal slogan is literally: "Be Evil."
[gnabgib]: Discussion (600 points, 11 days ago, 424 comments) https://news.ycombinator.com/item?id=50136881
[kaldev]: I felt like this whole thing is extremely inorganic. Funded by a WalMart heiress, website went up hours before a massive NYT ad buy, and (in my opinion) what felt like lot of coordinated / paid / botted engagement. Idk, I don't want the internet to be dictated by followers/engagement that can be easily botted via a SMM panel for pennies.
[_s_a_m_]: Zuckerberg Mr AIDS himself
Dayjob: Tax – evaluating AI agents on 50 professional tax assignments
landonxi · 10-28 15:46 UTC
该文介绍一项基准，用50个专业税务任务评估AI智能体的表现。
AI 判断：这属于专业基准评测，普通用户无法直接使用，也缺少可演示的消费级场景。
展开原始摘录
--- Top Comments ---
[lexandstuff]: I prepared my taxes this year using Astra + Codex. It did a thorough job on a complicated return, and I sent it to my accountant to fill out the tax form, with everything laid out in a spreadsheet. They sent it back for final review, and I had Codex check it; it was riddled with errors. The refund was nearly $2k lower than it should have been, and they put a bunch of stuff into the wrong fields. Had to go back and forth multiple times to get it 100% correct. Made me wonder how many previous y...
[bluejay2387]: Is this just giving a general-purpose agent a workspace full of documents and an assignment, then seeing what happens? I feel this doesn't really say anything. It appears they mostly dumped all of this in base models with little to nothing in the way of infrastructure to support it. If you just randomly grab some person off the street and say 'do taxes' they would probably mess up just as badly. The interesting question would be what the models could do in a well engineered har...
[RGS1811]: As a proof-of-concept, I've been using extra tokens to formalize the IRS tax forms and worksheets, mostly just to prove this can be done. The idea is that something like this could be adapted into a publicly maintained version of the IRS Direct File program (RIP). I'm up to an estimated 60% coverage of US federal tax situations so far. https://github.com/emmilco/us-tax-model
[waiwai933]: I have a non-trivial tax situation, as a US citizen living abroad. I prepare my own tax returns, never having found a cross-border tax preparer I trust, and have for the past few years, fed my draft return along with all the supporting paperwork into Claude. It has: 1. found errors, e.g. where I used the incorrect exchange rate (and managed to work out how I had made that error, which I was impressed by) 2. found items which would appear to be errors except for context it did not know 3. flag...
[holdingunsteady]: My tax accountant mis-filed my taxes a couple years ago, which cost me about $20K (which I learned after I switched to another accountant later). Don't know if Claude / GPT would make the same mistake or not given SOTA today, but feels like there's an opportunity for models to be saving me thousands or tens of thousands of dollars a year.
consumer451 · 10-28 15:31 UTC
Archive.today links going forward
AI 判断：前置规则过滤：HN 标题及已抓取内容未命中 AI 主题
展开原始摘录
--- Top Comments ---
[consumer451]: Archive.today = archive.is - which is added via comment on a huge percentage of posts on HN. Link to comment explaining the discovery: https://old.reddit.com/r/Seattle/comments/1xhf6qp/archivetod... edit: Maybe I should have titled this " claim to have discovered" as there is no external verification, and no write up. I thought I should post this in any case.
[lxe]: Again? This happened before, if you remember this: https://www.tomshardware.com/software/search-engines/archive... Previous HN thread: https://news.ycombinator.com/item?id=46624740
[surround]: If you're brave enough to want to see it yourself, visit https://archive.ph/jXv3A#:~:text=filed%20on%20Tuesday then click on the "Filed on Tuesday" link. It redirected me to https://stolidityseashellshotter[.]com/ezrsb5dd1268e6f6ebaad... which then redirected me to a fake clone of RT news: https://rurtnews[.]com/ Unlike the Reddit mods, I'm not getting redirected to the malware site, just the fake RT site. Edit: I clicked ...
[avidruntime]: Not sure what to believe here. Wouldn't it be possible to create a site that has some js embedded that is able to be aware of archive's process so that the archived page renders a different page than the original? Could that secondary page be what is trying to deliver the malware, using archive as an unwitting laundry? That matches the MO of malware delivery via other services, no? I'm baffled as to why the mods don't provide any details whatsoever. There are more links to...
[efficax]: They really need to show some evidence for this! Just this suggestion means I won't ever use it again
bschne · 10-28 14:38 UTC
CNC Stone Carving
AI 判断：前置规则过滤：HN 标题及已抓取内容未命中 AI 主题
展开原始摘录
--- Top Comments ---
[wolframhempel]: Before anyone complains that this removes the human aspect of art, it's worth stressing that sculptors have always looked for ways to automate work and save time (as the website itself points out). As far back as ancient Egypt, grids and measuring tools were used to reproduce figures at different scales. Roman workshops churned out copies of Greek statues in marble and bronze, using casts and moulds. From the 18th century onwards, sculptors used outright pointing machines to transfer mod...
[soci]: Worth noting that the stonemasons of Sagrada Familia (the catholic basilica Barcelona) have been using CNC to carve it's stones since early 1990s. Most Barcelonians thought they would never see the temple finished in their lifetimes, myself included. Part of the progress that made its completion possible this year is thanks to this technology improving over time. EDIT: Actually, it’s a mix of CNC roughing to shape the stone, followed by manual finishing. According to the Sagrada Família’...
[shellfishgene]: Wow, the monks are not messing around either, they built a full monastery with cathedral and all in gothic style!
https://newmountcarmelfoundation.org/building-a-mountain-mon...
[ricardobeat]: I love this story. They started roasting coffee and saving money for nearly a decade, until they could buy land in the mountains, and then set out to build a medieval monastery entirely in stone and developing the CNC machining. It's amazing what people can do when they have a goal.
[weirdish]: I've always wondered why we haven't seen a broad architectural movement returning to highly ornate decorative facades now that technology has made this kind of thing (more) possible. We could be having so much more fun in architecture right now!
bradley_taunt · 10-28 14:30 UTC
AlpineBSD 项目的 Wiki 页面，属于操作系统工程内容。
AI 判断：与普通人的 AI 新功能、工具或生活应用无关，按技术类低分处理。
展开原始摘录
--- Top Comments ---
[iamnothere]: Interesting to watch the rapid fragmentation of FOSS into “hard” anti-AI, AI-neutral-to-discouraged, AI-supplemented, and AI-driven camps, akin to the early GPL vs BSD wars. Suddenly I don’t see as many complaints about licensing, and some of the die-hard GPL stans are silently dropping that position as Linux adopts AI-supplemented development. On one hand it’s unfortunate, because it hypothetically dilutes developer efforts, but on the other hand the most extreme groups already had incompati...
[kev009]: These types of projects tend to have a flash of appeal, especially to people used to Linux, because it lessens one degree of unfamiliarity, but they don't seem to stick around for to me obvious reasons. It's hard to get involved with something like a BSD project because you have to learn a lot and put in enough effort to get committer rights. And then if you want to move the needle you have to do even more work and build consensus without getting overly emotional when people disagr...
[yjftsjthsd-h]: > AlpineBSD is not a fork of Alpine Linux, its package collection is original (though many packages are ultimately based on aports). > Alpine Linux uses the OpenRC init system. Currently, AlpineBSD uses s6 and we intend to add OpenRC as an alternative. I'm not sure where exactly the line should be, but that feels like it's not an Alpine* then. Like, if you're starting your package collection separately and use a different service manager, you're further from Alpine L...
[hoppp]: The whole "as little AI code as possible" is tribalism. I want software that is brilliant and well audited, no matter who or what produced it. No AI guarantees nothing, especially if it's a C codebase like netbsd, why not also join the rust tribe then?
[thenthenthen]: Alpine is also the codename of the first iOS, hence the default root pwd ‘alpine’.
https://mashable.com/archive/apple-ios-codenames#9.CLuUuywuq...
The October inflection point for vibe game development
Simon Willison · 10-28 13:55 UTC
这篇文章讨论“氛围式游戏开发”在十月出现的转折点。
AI 判断：标题初筛（未依据正文）：仅凭标题看，这是面向开发者的趋势评论，缺少普通人可直接使用或可演示的具体功能，暂不适合作为大众教程选题。
展开原始摘录
I've written before about how November 2025 was
an inflection point for coding agents
- the point where Opus 4.5 and GPT-5.1 crossed a tipping point that meant their respective coding agents became reliable daily drivers for software engineers. I think we've reached a new inflection point in October 2026, and this time it's for
vibe coding computer games
.
Tree Appreciator
Emberwick
Battle Drones
Pumpkin Farmer, aka Quarter Sun Cove
Four times as insulting
What happens if someone actually puts the work in?
You should build a game!
Claude Opus 5.5 in particular exhibits meaningful improvements in several areas needed to build a compelling game:
3D modeling
Music generation (see
Scrimshaw Jukebox
)
Game design (!)
That last one is particularly important. Models have been able to create something that
looks
like a game for quite a while now, as seen in
my previous experiments
. The challenge has been getting to a game that's actually
fun to play
.
Game design is a deep subject area. I won't pretend to be well versed in it - I know enough to have respect for it as a craft. Creating a novel, compelling gameplay loop which challenges players at the right level and provides a satisfying learning curve is a fine art. I didn't think the models could do it.
Then I tried telling Claude Opus 5.5 to "Make it a LOT more fun"...
and it did
.
In the past few weeks I've vibe coded (in the
classic sense of the term
- no looking at the code at all) four games that I've genuinely enjoyed playing.
I attended
Foo Camp
in Berkeley over the weekend and met several people who were having similar experiences. Building games that play well is shockingly easy now. The new challenge is coming up with concepts for games that are novel and delightful and that wouldn't exist without AI assistance.
I haven't done an exhaustive study, but my current impression is that GPT-6 Astra isn't
quite
there yet. It can absolutely build a game, but its sense of fun doesn't match that of Opus 5.5. I expect OpenAI will get there by the end of the year.
Here are the four game projects that have convinced me that the models are now capable of producing fun games with very little guidance. All four are built in HTML and JavaScript and run entirely in the browser.
Tree Appreciator
The smallest game I found to be genuinely delightful is
Tree Appreciator
. I used the regular Claude chat interface for this - it didn't even need Claude Code.
I built this on a whim while we were driving along
Skyline Boulevard
looking at trees. My prompt was:
Build a three js game called "Tree appreciator" - it shows you two algorithmically generated trees and the game mechanics is about appreciating the differences between them. Think hard about how to make sure it is both fun and beautiful.
I've found that telling models to use
three.js
works really well. Claude doesn't have its own image generation model but is capable of building full-color 3D models that look and feel excellent.
This game required no iteration or follow-up prompting at all - I got the game in a single shot. It gave me back a charming little mobile-friendly spot-the-difference game. You can
play that here
. Here's
the chat transcript
that built it.
It's pleasingly relaxing. The graphics are fun, and collecting trees that you have completed in a "herbarium" provides a nice, cosy sense of accomplishment.
Emberwick
This was a more ambitious project that I again built using regular Claude chat. I started out by building an experimental prototype of an engine for a top-down fantasy RPG:
Build a demo game engine three.js which is a top down (but at an angle) RPG - the player gets to move through the world with one or two companions. Create a fantasy village setting that demonstrates walking around outside and then shows how when the player enters a building the wall between the player and the camera goes transparent so the user can see what is happening
Add some NPCs and plenty of furniture and different types of building some with an upstairs - be creative, make it look great
Here's
that first version
. It worked really well!
I was inspired to increase my ambitions:
Make it not so flat - the buildings are great but the landscape itself is completely flat at the moment, fix that
Add basements to two of the buildings with a tunnel system joining them below the village
Add the start of a rocky mountain on one side of the village containing a cave with a sleeping troll
Here's
that version
.
When
that
produced a good result, I decided to push it all the way:
Now turn this into an actual game - quadruple the size of the playable area, introduce fun game mechanics, add a plot that isn't too cliche, some twists, a magic system, make it really good
We're back to prompts like "make it really good" here! Asking for a plot that "isn't too cliche" seemed to work too - it's a bit cliché, but still novel enough that I enjoyed following the story.
Unlike Tree Appreciator, this took quite a bit of additional playtesting and prompting to get to a finished version. This was mainly due to gameplay problems - things like a character not being in the right place to complete a quest because a different story element had caused them to move to another location.
Here's
that final game
. It has a full story that takes about 15 minutes to play through.
There's even an experimental level editor if you want to drop in your own houses, furniture, and landscape features.
Once it was all working (and I'd iterated through some gameplay bugs), I played it through from start to finish and found myself emotionally affected by the ending! I'm somewhat embarrassed by this, because
everything
in this game - the graphics, the plot, the characters, the music, the dialog - came from the model. My creative input was pretty much just "it's a fantasy village and there's a troll asleep in a cave".
Here's
the full shared Claude chat
in which I built this game.
I also fed that same initial sequence of prompts through ChatGPT GPT-6 Pro. Here's
the transcript
and
the finished game
for comparison. It looks great, but I don't think it plays nearly as well.
Battle Drones
This is by far my most successful project, which I
wrote about last week
. It's a Command & Conquer-style Real-Time Strategy game where two robot armies (or four
as of Saturday
) build units, harvest scrap, and battle over territory.
This is also the project that convinced me that Opus 5.5 knows what "fun" is.
The first versions were built back in July using Claude Fable 5. It got me to a workable 2D Command & Conquer-style game, but it wasn't great to play.
I prompted Opus 5.5 to "Make it a LOT more fun" and then later to "Port this game to three.js" and the game was transformed. There's a blow-by-blow account (prompts included) in
my write-up
.
Today, Battle Drones has 2-4 player multiplayer (via a vibe coded WebSockets relay server), medals, replays, seven maps, a random map generator, and a set of "mutators" that add variants of the rules.
It also lets different LLMs play against each other, using a mode where the game advances five seconds and then pauses to give different models time to interpret the new game state and line up their next set of commands. Here's a replay of
Claude Haiku 5.5 vs. GPT-6 Luna
:
Battle Drones is a particularly good example of Opus 5.5's ability to create 3D models. This page lets you
browse all of the models in the game
.
I released Battle Drones under the Apache 2.0 License last Tuesday. On Wednesday Gabriel Turpin released
Panache
(
screenshots here
), a fork of Battle Drones that reimagined the game for steampunk 1900s Belle Époque France, with new models and music and an animated intro sequence. Open source at work!
Pumpkin Farmer, aka Quarter Sun Cove
I live in
Half Moon Bay, California
, home of the annual
Half Moon Bay Art and Pumpkin Festival
, which includes a giant pumpkin weigh-off competition.
I decided to try building a game that wasn't just a cheap imitation of something that already existed, so I set out to create a game about farming pumpkins, growing giant pumpkins, and handling the rush of the October pumpkin season.
I called it "Quarter Sun Cove".
I'm still iterating on this project, but it's shaping up quite nicely. It's a cosy farming simulator where you spend the year tending to your giant pumpkin and planting out regular-sized pumpkins for sale in October, then sell pumpkins during pumpkin season and enter your giant pumpkin in the weigh-off.
Every character in it is a (vegetarian) animal. The mayor of the town is a pelican. You can be a pelican, skunk, raccoon, possum, beaver, coyote, fox, owl, sea otter, or harbor seal. Those species all later show up (accompanied by their children) to buy pumpkins from you.
It's
absurdly
cute.
You can
play the latest version here
. I'm not convinced by the game mechanics yet - it's actually quite complicated: you have to decide when to plant pumpkins such that they will become ripe at the weekends in October with the most visitors.
Here's the
initial brainstorming session
, which resulted in
this spec file
. Then I had Opus 5.5 Ultracode burn millions of tokens and nearly two days building it.
I told Claude to "surprise and delight the user frequently" and the result is
so many
little details and easter eggs. There are mischievous raccoons, an elephant seal that runs the weigh-off, harbor seals that roll their pumpkins back into the ocean, a wandering goat, and superstitions about growing giant pumpkins that
sometimes
come true.
This isn't my best game yet - it's not as fun to play as Battle Drones - but it represents the kind of game I want more of: weird, fun, uniquely personal - something that wouldn't have been built in the before-times.
Four times as insulting
If you are offended by AI-generated art, AI-generated music, AI-generated writing, and AI-generated code, I have bad news for you: vibe coded games combine
all of those at once
.
The wider games community already has an aversion to AI-generated assets in games. This has been a growing issue over the past few years as tools like Midjourney have enabled developers to generate assets for their projects.
This IGN story
covering complaints about slop in Valve's Steam Next Fest is a good recent example.
We now have games where
everything
- code, assets, dialog, and even the game design itself - can be outsourced to the models. And the results don't suck!
Building a computer game used to be one of the most challenging projects in all of software development, combining a diverse range of skills and requiring intense focus and effort over many months or even years. And now here I am spitting out a playable multiplayer RTS by prompting on my phone!
I expect that, as we've already seen with programming, this will result in a split. Some players will refuse to have anything to do with games that were built in this way. Others may embrace this as opening up game development to a much wider audience. I lean in that direction!
I certainly won't be entering any of my vibe coded games in traditional game jams or competitions - I don't feel like I've earned that right. We'll likely see new avenues emerge for vibe game developers to compare their projects.
What happens if someone actually puts the work in?
The thing I find both exciting and shameful about this is that I have got interesting results from
so little creative input
. The projects I've described here are loosely ordered by how much effort I've put into them. Not a single project represents more than four hours of my work (plus many more hours of Claude Code churning away by itself).
Now imagine what could be achieved by anyone who takes this seriously, invests significant creative time, applies better taste in game design, and then playtests and iterates on the result until it's as good as they can make it.
I want to see games that wouldn't have existed without these new tools.
You should build a game!
Building games in this way is
so much fun
. It's the perfect application for vibe coding because it's so low stakes: how much damage can you cause with a bug in a personal vibe coded game project?
(OK, you can do some damage... my attempt at audio-based multiplayer gave my wife Natalie
a nasty headache
!)
All of my games so far have been HTML and JavaScript using static hosting (generally GitHub Pages or Cloudflare Pages). Modern browsers - including mobile browsers - are incredibly capable. Frontier models already know libraries like
three.js
and
Phaser
. I'm finding these mobile web games to be indistinguishable from the native experience, at least for the kinds of games I'm trying to build.
I suggest starting with a prompt like this:
Build a mobile-friendly three.js game where...
This works on both
chatgpt.com
and
claude.ai
- you don't even need to fire up a full coding agent. As one last example, here's "Build a mobile-friendly three.js game where a mouse has to hide from a cat"
by Claude Opus 5.5 Extra
and
by GPT-6 Extra High
.
Switch to Codex or Claude Code once your project passes the proof-of-concept phase.
And if the game isn't fun enough, add this:
Make it a LOT more fun
Tags:
game-design
,
games
,
ai
,
generative-ai
,
chatgpt
,
llms
,
claude
,
vibe-coding
,
claude-code
Kentucky man becomes the first humanist granted religious exemption from AI
danso · 10-28 13:41 UTC
美国肯塔基州一名人文主义者以宗教理由获准不使用 AI，成为据报道的首例此类豁免。
AI 判断：这则新闻有社会讨论价值，但个案适用范围和先例效力仍需核实，且缺少可演示的实用效果。
展开原始摘录
--- Top Comments ---
[ahmedfromtunis]: This is very interesting. As someone who thinks AI is an amazing tool, I am very well interested in seeing how will people refusing to use AI fare both in professional settings and day-to-day life. A new Amish-like lifestyle may be developing before our eyes in real-time and that's very interesting to follow and observe (from a distance).
[smag]: The revenge effects of technologies are well known: email was great! It was faster and much cheaper than mailing letters. But it gave rise to spam and many types of fraud. The mobile phone is awesome. A powerful computer in your pocket. But now people, including teenagers, spend 12-16 hours each day staring into their mobile phones, doomscrolling towards Babylon. In 2006 people memorized dozens of phone numbers. Today? Maybe you remember 1-2 numbers. What will be the revenge effects of LLMs a...
[6gvONxR4sf7o]: It's interesting to think about this expanded view of religion and how it interacts with nondiscrimination laws. The guy's beliefs (at least in the quotes in the article) seem like a good combination of well reasoned and sensible and core faith-like beliefs, so I'm glad they're protected. At the same time, I can imagine plenty of beliefs that would make me not want to hire someone. I know this one didn't go to court, but I wonder what tests courts use to draw the line...
[hereme888]: I've had the concern for years that AI users (or AI itself) will eventually consider non-AI users a pain, because humans will become inefficient in lots of workloads and management, sometimes causing things to come down to a screeching halt, relatively speaking. We have to solidify laws protecting anti-AI-ers if that's their choice.
[throwfaraway135]: > This invention, O king, will make the Egyptians wiser and will improve their memories; for it is an elixir of memory and wisdom that I have discovered. Unsurprisingly history repeats itself, there will always be people who reject the unknown for the comforts of the past.
Emacs Turns 50
internet_points · 10-28 13:25 UTC
《麻省理工科技评论》发文回顾经典文本编辑器 Emacs 诞生 50 年的发展历程与社区文化。
AI 判断：这是一篇关于程序员编辑器的纪念性文章，与普通人的 AI 使用无关，也缺乏可演示、可上手的实用价值。
展开原始摘录
--- Top Comments ---
[jwr]: I have been using Emacs for more than 30 years now. Other editors have come and gone, Emacs stayed. Parts of my configuration are more than 30 years old. There were better and worse times in Emacs history. Recent times, however, are really good. Emacs is maintained and developed actively and there are many external packages which are impressive. Magit and ghostel are good examples of packages that I use on a daily basis to get my work done. Recently, AI completely changed how I use Emacs, and...
[RedCinnabar]: I programmed in Emacs in a myriad of different languages, I used it to write my thesis, to write my homework and later to grade them. I’ve used it to publish countless blog articles, wiki pages and technical documents. I read that I was hired at my first job from en email received inside Emacs, I’ve used it to send countless emails to colleagues, employers and business partners. I planned both my personal and professional life inside it, I’ve used it to track the time spent doing a certain ta...
[iLemming]: 50 years old and feels more modern than any of the crap folks try to sell you "as the editor of the future". Emacs is the true editor of the past, present and distant future. Modal editing was supposed to pull users away - Evil mode made vimming better. LSPs were supposed to kill it - only made it better. Electron-based crap was supposed to make terminal-native editors look ancient with a built-in terminal, extensions, and a polished UI. Emacs got vterm, eat, and ghostel. GitKraken-...
[marttt]: Is anybody here consciously trying to keep their Emacs configs/setup "lean", minimalist, as small as possible? I'm not a coder, but I did use Emacs daily around 2010-2012, with brief returns later. I always gave up eventually because the complexity of addons, the org-mode-can-do-everything-so-I-should-do-everything-with-it etc just grew over my head. I've discovered that I like Plan9 Sam, Acme and, in general, the one-thing-well camp better. Each their own, I guess. T...
[osener]: After a couple of years of detour I’m back to using Emacs since all the fancy lsp, debugger etc stuff in VSCode and Zed doesn’t matter as much for me anymore and I am just more productive with Emacs. But I am still using it like it is 2020. I hear people using coding agents to make crazy customizations to their emacs config and I can imagine Emacs is particularly suitable for this. Curious what are the biggest things that have come out in the last year or two, and what emacs based personal s...
OpenAI and ServiceNow expand partnership across enterprise workflows
OpenAI News · 10-28 13:15 UTC
OpenAI 与 ServiceNow 宣布扩大合作，将覆盖更多企业工作流程。
AI 判断：标题初筛（未依据正文）：标题仅显示企业合作扩展，缺少普通人可直接使用或演示的具体功能信息。
展开原始摘录
OpenAI and ServiceNow expand their partnership to bring OpenAI models and agents into ServiceNow and ServiceNow workflows into ChatGPT and Codex.
Conversational AI in banking: Use cases, benefits, examples
ElevenLabs Blog · 10-28 12:00 UTC
这篇文章介绍对话式 AI 在银行业的应用场景、优势与案例。
AI 判断：标题初筛（未依据正文）：标题仅为行业用例综述，缺少普通人可直接使用的新功能或可演示细节，时效性也较弱。
展开原始摘录
Learn what conversational AI in banking is, how voice agents handle phone calls, use cases, core metrics, and how banks and fintechs use ElevenAgents today
Lettuce cuts creative ad production from over a week to under a minute with ElevenLabs
ElevenLabs Blog · 10-28 12:00 UTC
ElevenLabs 博客称，Lettuce 使用其工具把创意广告制作周期从超过一周压缩到不到一分钟。
AI 判断：标题初筛（未依据正文）：标题仅为厂商案例宣传，缺乏可复现步骤和普通人可直接使用的信息，实用性有限。
展开原始摘录
Lettuce uses Ads Engine in ElevenCreative to generate Meta ad variants in 45 seconds, cutting creative costs by 75% and cost per lead by 40-60%.
ElevenLabs and adELA bring free voice restoration to Spain
ElevenLabs Blog · 10-28 12:00 UTC
ElevenLabs 与 adELA 合作，在西班牙为失语人群免费提供声音恢复服务。
AI 判断：标题初筛（未依据正文）：仅凭标题判断，这是面向失语患者的免费声音恢复服务落地西班牙，受益人群真实但范围较窄，缺少价格与使用细节。
展开原始摘录
ElevenLabs partners with adELA, Spain’s leading ALS association, to restore the voices of patients living with ALS for free.
How your transcript knows where each word is spoken
ElevenLabs Blog · 10-28 12:00 UTC
ElevenLabs 介绍其转录文本如何定位每个词被说出的位置。
AI 判断：标题初筛（未依据正文）：标题仅指向语音转录对齐的技术机制，缺少普通人可直接使用的新功能或可演示场景。
展开原始摘录
Meet Cue, ElevenLabs’ forced alignment model: character-level timestamps, any recording length, and an hour of audio aligned in under a second.
Voice AI for Karnataka
ElevenLabs Blog · 10-28 11:30 UTC
ElevenLabs 博客介绍面向印度卡纳塔克邦的语音 AI。
AI 判断：标题初筛（未依据正文）：仅凭标题看，这是地区性语音 AI 案例，对中文普通用户的实用性和可演示性尚不明确。
展开原始摘录
ElevenLabs signs an MoU with the Government of Karnataka to bring voice AI to citizen services, voice AI safety and voice restoration.
galadran · 10-28 11:03 UTC
Apple's git still trusts the compromised eMudhra certificates
AI 判断：前置规则过滤：HN 标题及已抓取内容未命中 AI 主题
展开原始摘录
--- Top Comments ---
[chrisweekly]: Right, so it's not just git, it's a bunch of system-provided utilities; macOS trusts bad certs, and these utils that trust the OS are affected. I use homebrew for my git and curl but haven't yet checked to confirm whether that matters here.
Kind of outrageous that Apple's guidance is to edit the cert chain manually, and to expect that subsequent OS updates may revert your mitigation, reintroducing the vulnerability (as happened at least once so far).
[jrochkind1]: That MacOS hasn't removed from OS-level /etc/ssl seems like a big deal, affecting more than just OS-supplied git. Is /etc/certs not used by most graphical Mac software, I guess? The chaotic state of OS-level cert management, on Linux as well, is pretty dismaying. If Apple is still trying to brand themselves as security-forward, you'd think removing the revoked CA certs from /etc in a system update would be a pretty easy thing to do?
[truelson]: Here's what apple (themselves) recommends to patch: https://support.apple.com/en-us/149555 Pretty much every developer on mac should probably run their patch.
[qaz_plm]: Apple’s statement: https://support.apple.com/en-us/149555
[kryptiskt]: I'm not being sarcastic when I say that Apple should copy Windows Update, I have no idea how they have got this far without a method to add small patches outside monolithic OS updates.
The Decoder：AI News（RSS） · 10-28 10:56 UTC
Cathedral 利用 AI 智能体侵入 Anthropic 内部系统，但双方对事件涉及范围的说法不一致。
AI 判断：标题初筛（未依据正文）：标题显示这是 AI 安全事件，但普通用户的实用性和可演示性有限。
展开原始摘录
网络安全初创公司 Cathedral 通过第三方业务软件渗入 Anthropic 内部系统，称其 AI 智能体在不到十二小时内接触到会议记录、薪资数据和 Slack 消息，并借漏洞赏金计划披露问题。
爱范儿 · 10-28 10:21 UTC
据标题所述，有一款机器人可以从人的嘴部进入体内，用途是医疗救治。
AI 判断：标题初筛（未依据正文）：这个标题呈现了画面感强、很新奇的医疗机器人场景，但仅有标题可供判断，普通人能否实际使用以及事实细节都还无法核实。
展开原始摘录
当所有人都在讨论 AI 时，机械工程同样可以给世界带来变化。
#欢迎关注爱范儿官方微信公众号：爱范儿（微信号：ifanr），更多精彩内容第一时间为您奉上。
Quoting Terk Terkowitz
Simon Willison · 10-28 10:09 UTC
该条目仅是一则引用特克·特科维茨言论的短文，标题未说明具体内容。
AI 判断：标题初筛（未依据正文）：仅有标题且为他人语录引用，无任何可核实的信息或可用功能，无法支撑面向普通人的选题。
展开原始摘录
If your work does not benefit from current AI tooling, that is a finding, not a failing. Document why, share it with your community of peers, bring it to your manager, and work to find solutions. An engineer who can explain where AI does not yet help their work is demonstrating the judgement we are asking for.
—
Terk Terkowitz
, An update from Amazon engineering leadership on the role of the developer in the age of AI
Tags:
amazon
,
generative-ai
,
agentic-engineering
,
ai
,
llms
Google’s AI worlds can now talk back
The Rundown AI · 10-28 10:00 UTC
标题称谷歌的 AI 世界如今能够“回话”了。
AI 判断：标题初筛（未依据正文）：仅凭标题难以确认具体功能、可用地区和演示效果，实用性与受众范围仍有不确定性。
展开原始摘录
PLUS: Make editable product mockups with GPT-6.1 Terra
The future of the firm
OpenAI News · 10-28 10:00 UTC
OpenAI 新闻发布题为《公司的未来》的文章，围绕企业未来展开讨论。
AI 判断：标题初筛（未依据正文）：仅凭标题判断，这是一篇观点性文章，缺少可演示的具体功能或操作步骤，实用性和画面感有限。
展开原始摘录
How AI might enter the org chart—beneath, beside, and between us.
X：Saito 硬地骇客 (@SaitoWu) · 10-28 09:32 UTC
据标题信息，Anthropic 的 S-1 文件显示 90% 增量毛利率，SemiAnalysis 播客对其中数字进行了解读。
AI 判断：标题初筛（未依据正文）：这是基于标题的财务与播客解读，对普通人的实用性和画面感较弱，且仅有标题信息可供核实。
展开原始摘录
Saito 转述 SemiAnalysis 播客对 Anthropic S-1 的解读：招股书口径为增量毛利 90%、推理毛利 85%、连续两个季度盈利，Q1 净收入留存达 500%，收入同比涨 1368% 而算力费用只涨 300%。
Legora is building the agentic operating system for legal work
OpenAI News · 10-28 09:30 UTC
Legora 正在打造面向法律工作的智能体操作系统。
AI 判断：标题初筛（未依据正文）：基于标题判断，这是法律行业的智能体平台消息，缺少普通人可用的功能、价格或演示信息，实用性有限。
展开原始摘录
See how Legora uses OpenAI models, ChatGPT, and Codex to accelerate legal workflows, expand contract analysis, and free lawyers to focus on higher-value work.
AIBase 中文 · 10-28 09:23 UTC
据媒体报道，OpenAI计划向谷歌采购TPU芯片，并招募工程师进行模型适配，被视为其算力供应格局可能出现的新变化。
AI 判断：这是基于媒体报道的芯片采购与算力供应链消息，属于产业与基础设施层面，普通用户既无法使用也没有可演示的效果，且消息本身尚未被官方证实。
展开原始摘录
AI资讯 AI新闻资讯 正文 OpenAI被曝将向谷歌采购TPU芯片，算力供应链格局再变 发布于AI新闻资讯 发布时间 : 2026年10月28号 17:23 阅读 : 1 分钟 10月28日消息，据媒体报道，作为全球人工智能领域的头部玩家，OpenAI正计划向谷歌采购人工智能芯片。长期以来，OpenAI一直是英伟达图形处理器（GPU）最重要的买家之一，主要依赖其芯片来完成海量AI模型的训练与复杂的推理运算。 随着行业对算力的需求持续飙升，企业正积极寻求更具性价比且适合运行AI模型的替代方案。谷歌自研的TPU（张量处理器）近年来逐渐崭露头角，并在今年开始向部分外部客户直接销售。回顾双方合作的演进历程，在过去16个月里完成了清晰的“三跳”:2025年6月，OpenAI 首次 开始租用谷歌的算力;同年7月，其公开表示正对谷歌TPU开展早期测试，但暂无大规模部署计划;而到了如今，双方正式迈向了实质性的采购计划。此外，OpenAI目前也正在积极招募负责将其模型在TPU上进行适配与优化的工程师，而模型层的底层适配通常正是大规模采购落地的重要先行指标。 回顾其算力拓展轨迹，谷歌TPU曾在OpenAI创业早期发挥过重要作用，此后OpenAI全面转向英伟达生态，并陆续接入了包括AMD、Cerebras以及AWS等在内的多家供应商芯片。 对于行业格局而言，这一转变释放出了明显的信号:随着其最重量级的客户在单一GPU路线之外开辟新的算力分支，英伟达订单的独占性正逐渐减弱。 权威 机构TrendForce的数据显示，英伟达在AI服务器处理器市场的份额预计将从2025年的69.6%下滑至2026年的64%，整体行业趋势与当前的技术和供应链多元化布局高度契合。 相关推荐 豆包公关回应"15 万群演被喂模型"：传言严重失实 豆包公关负责人刘星发文辟谣，称“横店与字节合作、将15万名群演肖像和声音喂给字节大模型”严重失实。他介绍，真正的“横店影视AI服务平台”旨在建设公共数字基础设施，将影视基地实景等资产整理为高质量数据集，帮助剧组高效选景、筹备和拍摄，并降低中小影视企业使用AI的门槛，让更多创作者有条件做好作品。 2026年10月28号 17:13 148.0k 外国博主用AI耳机当看诊翻译:“直接听、直接翻，整个交流会顺很多” 外国博主“Carlo吴（卡罗）”在社交平台分享看中医经历。他带上千问AI耳夹式耳机，与医生各戴一只，用面对面翻译辅助交流。卡罗称自己觉得传统医学很有意思，但语言始终是障碍。看诊时医生问哪里不舒服，他只能边比画边说“全部不舒服”；戴上耳机后，两人开始进一步沟通。他感叹这是第一次真正听懂中医的话。 2026年10月28号 16:54 160.1k CVE 漏洞数近乎垂直拉升，AI 成安全研究"放大器" 科技媒体gigazine报道，开发者@0x_Osprey推出网页版CVE Tracker，将已获CVE编号的漏洞数量变化制成可视化图表，显示2026年以来曲线近乎垂直上升。CVE是全球通用漏洞编号体系，安全研究员、厂商和企业靠它追踪、通报和修复缺陷。该工具整理公开记录成趋势图，并叠加主流AI模型发布时间，便于直观对照漏洞增长与AI发布。 2026年10月28号 16:25 164.8k 百度世界大会前夕重磅流出：消息称其将披露世界模型进展 10月28日，《科创板日报》独家报道，百度计划近日正式披露“世界模型”研发进展与突破性成果。该领域是全球AI最火热前沿之一，核心技术是对真实物理环境精准动态建模与未来预测。业内分析认为，若该技术率先落地自动驾驶，将大幅提升无人车在复杂路况下的场景预测与智能决策水平。 2026年10月28号 16:07 166.7k Liquid AI重磅发布LFM3 多模态模型家族，主打高效专用智能 近日，Liquid AI正式发布新一代基础模型家族LFM3。首个版本包含六款约2B参数模型，覆盖文本、视觉、音频、全模态与决策等维度，均基于LFM3-1.7B-Base，预训练吞吐达22T Token。该系列采用全新三混合层架构，融合门控短卷积、滑动等机制，主打参数更少、效率更高。 2026年10月28号 16:04 153.6k 智启未来，您的人工智能解决方案智库 简体中文 © 2026 AIBase 备案号：闽ICP备08105208号-24
CVE 漏洞数近乎垂直拉升，AI 成安全研究"放大器"
AIBase 中文 · 10-28 08:25 UTC
开发者制作的网页版 CVE Tracker 把 2026 年以来已获编号的漏洞数量画成近乎垂直上升的曲线，并叠加主流 AI 模型发布时间，报道认为 AI 只是加快了扫描与上报效率，发现变多并不等于系统更危险。
AI 判断：这是安全研究和漏洞统计层面的技术话题，与普通人的日常使用关联很弱，也没有可亲自演示的消费级功能，因此只能给低分。
展开原始摘录
AI资讯 AI新闻资讯 正文 CVE 漏洞数近乎垂直拉升，AI 成安全研究"放大器" 发布于AI新闻资讯 发布时间 : 2026年10月28号 16:25 阅读 : 1 分钟 科技媒体 gigazine 当日发文称，开发者 @0x_Osprey 做了一张网页版 CVE Tracker，把已经拿到 CVE 编号的漏洞数量变化画成可视化图表，清晰呈现出 2026 年以来那条近乎垂直上升的曲线。 CVE 是全球通用的漏洞编号体系，安全研究员、软件厂商和企业都靠它来追踪、通报和修复缺陷。CVE Tracker 把这些公开记录整理成趋势图，还贴心地把主流 AI 模型的发布时间叠了上去，让人一眼看清漏洞发现高峰与模型问世之间是否存在呼应。（该站点目前只有深色界面。） 值得注意的是，AI 公司常常在模型正式发布前，就用测试版本提前挖洞、报洞，所以模型上线时刻，未必正好踩在漏洞激增的节点上。 报告变多，不等于系统更危险 gigazine 的判断是：AI 正充当安全研究的"放大器"——它更快速地扫描代码、复现异常、整理成报告，把人力原本顾不上的软件角落也拉进排查范围。效率被抬高的同时，发现量自然水涨船高。 但报道也泼了盆冷水：漏洞"被发现得更多"，并不等同于系统"变得更危险"。更多的上报，既可能揭开真实风险，也可能夹带低质量、重复或难以验证的噪声，反过来加重开源维护者与安全团队的分诊负担。当 AI 把发现漏洞的门槛摊平，如何甄别真伪、避免被海量报告淹没，成了新的考题。 相关推荐 豆包公关回应"15 万群演被喂模型"：传言严重失实 豆包公关负责人刘星发文辟谣，称“横店与字节合作、将15万名群演肖像和声音喂给字节大模型”严重失实。他介绍，真正的“横店影视AI服务平台”旨在建设公共数字基础设施，将影视基地实景等资产整理为高质量数据集，帮助剧组高效选景、筹备和拍摄，并降低中小影视企业使用AI的门槛，让更多创作者有条件做好作品。 2026年10月28号 17:13 148.0k 外国博主用AI耳机当看诊翻译:“直接听、直接翻，整个交流会顺很多” 外国博主“Carlo吴（卡罗）”在社交平台分享看中医经历。他带上千问AI耳夹式耳机，与医生各戴一只，用面对面翻译辅助交流。卡罗称自己觉得传统医学很有意思，但语言始终是障碍。看诊时医生问哪里不舒服，他只能边比画边说“全部不舒服”；戴上耳机后，两人开始进一步沟通。他感叹这是第一次真正听懂中医的话。 2026年10月28号 16:54 160.1k 百度世界大会前夕重磅流出：消息称其将披露世界模型进展 10月28日，《科创板日报》独家报道，百度计划近日正式披露“世界模型”研发进展与突破性成果。该领域是全球AI最火热前沿之一，核心技术是对真实物理环境精准动态建模与未来预测。业内分析认为，若该技术率先落地自动驾驶，将大幅提升无人车在复杂路况下的场景预测与智能决策水平。 2026年10月28号 16:07 166.7k Liquid AI重磅发布LFM3 多模态模型家族，主打高效专用智能 近日，Liquid AI正式发布新一代基础模型家族LFM3。首个版本包含六款约2B参数模型，覆盖文本、视觉、音频、全模态与决策等维度，均基于LFM3-1.7B-Base，预训练吞吐达22T Token。该系列采用全新三混合层架构，融合门控短卷积、滑动等机制，主打参数更少、效率更高。 2026年10月28号 16:04 153.6k 腾讯WorkBuddy上线「远程控制」：手机可实时观看并接管电脑桌面，远程操作不消耗Token 10月28日，腾讯WorkBuddy上线“远程控制”，由腾讯Edge One团队支持。手机App一键连电脑，实时看并接管桌面；画面以小窗悬浮首页或会话页顶部，输入框可用，任务跑偏可纠正。全屏横屏后，手指点哪即点击电脑同位置；右侧有“操作”“键盘”“鼠标”工具，“操作”可平铺窗口、返回桌面、切换画面等。 2026年10月28号 15:39 178.8k 智启未来，您的人工智能解决方案智库 简体中文 © 2026 AIBase 备案号：闽ICP备08105208号-24
AIBase 中文 · 10-28 08:07 UTC
百度世界大会前有消息称其将披露世界模型研发进展，但具体商业用途及与萝卜快跑的关联尚未明确。
AI 判断：该消息属于大会前传闻，缺少可演示功能、使用限制和普通人可操作信息，更适合技术观察而非大众教程。
展开原始摘录
AI资讯 AI新闻资讯 正文 百度世界大会前夕重磅流出：消息称其将披露世界模型进展 发布于AI新闻资讯 发布时间 : 2026年10月28号 16:07 阅读 : 1 分钟 10月28日消息，据媒体报道，百度计划于近日正式对外披露其在“世界模型”领域的研发进展与突破性成果。作为当前全球人工智能领域最火热的前沿方向之一，世界模型的核心技术在于对真实物理环境进行精准的动态建模与未来预测。 业内分析指出，如果该技术能够率先落地并应用于自动驾驶系统，将大幅提升无人车在复杂路况下的场景预测能力与智能决策水平，从而为无人车的规模化商业落地提供更强大的底层支撑。在过去数年间，百度旗下的“萝卜快跑”自动驾驶业务已在全球28座城市落地运营，并积累了海量的真实道路运行数据，这也为其世界模型的研发与验证奠定了深厚的基础。不过，目前官方尚未明确公布该模型的具体商业用途以及与萝卜快跑业务的直接关联。 巧合的是，百度此前已官宣“百度世界2026”大会定档于11月12日在北京国家会议中心二期举办。本届大会以“下一幕，智能组织”为主题，呼应了李彦宏在十年前提出的“互联网的下一幕，是人工智能”这一经典判断，将全行业的目光进一步引向AI驱动的组织与产业深度重构。随着大会开幕日期的临近，这项备受关注的世界模型进展是否会在此次盛会上正式亮相，已成为科技界与资本市场瞩目的焦点。 相关推荐 OpenAI被曝将向谷歌采购TPU芯片，算力供应链格局再变 10月28日消息，OpenAI计划向谷歌采购人工智能芯片。OpenAI长期是英伟达GPU重要买家，依赖其完成大模型训练与推理。随着算力需求飙升，企业正寻求更具性价比、适合运行AI模型的替代方案。谷歌自研TPU近年逐渐崭露头角，成为潜在选择。此举或反映AI芯片采购多元化趋势。 2026年10月28号 17:23 137.7k 豆包公关回应"15 万群演被喂模型"：传言严重失实 豆包公关负责人刘星发文辟谣，称“横店与字节合作、将15万名群演肖像和声音喂给字节大模型”严重失实。他介绍，真正的“横店影视AI服务平台”旨在建设公共数字基础设施，将影视基地实景等资产整理为高质量数据集，帮助剧组高效选景、筹备和拍摄，并降低中小影视企业使用AI的门槛，让更多创作者有条件做好作品。 2026年10月28号 17:13 148.0k 外国博主用AI耳机当看诊翻译:“直接听、直接翻，整个交流会顺很多” 外国博主“Carlo吴（卡罗）”在社交平台分享看中医经历。他带上千问AI耳夹式耳机，与医生各戴一只，用面对面翻译辅助交流。卡罗称自己觉得传统医学很有意思，但语言始终是障碍。看诊时医生问哪里不舒服，他只能边比画边说“全部不舒服”；戴上耳机后，两人开始进一步沟通。他感叹这是第一次真正听懂中医的话。 2026年10月28号 16:54 160.1k CVE 漏洞数近乎垂直拉升，AI 成安全研究"放大器" 科技媒体gigazine报道，开发者@0x_Osprey推出网页版CVE Tracker，将已获CVE编号的漏洞数量变化制成可视化图表，显示2026年以来曲线近乎垂直上升。CVE是全球通用漏洞编号体系，安全研究员、厂商和企业靠它追踪、通报和修复缺陷。该工具整理公开记录成趋势图，并叠加主流AI模型发布时间，便于直观对照漏洞增长与AI发布。 2026年10月28号 16:25 164.8k Liquid AI重磅发布LFM3 多模态模型家族，主打高效专用智能 近日，Liquid AI正式发布新一代基础模型家族LFM3。首个版本包含六款约2B参数模型，覆盖文本、视觉、音频、全模态与决策等维度，均基于LFM3-1.7B-Base，预训练吞吐达22T Token。该系列采用全新三混合层架构，融合门控短卷积、滑动等机制，主打参数更少、效率更高。 2026年10月28号 16:04 153.6k 智启未来，您的人工智能解决方案智库 简体中文 © 2026 AIBase 备案号：闽ICP备08105208号-24
Thibault Sottiaux: And it's done. See you tomorrow for some good news. https://t.co/WhJle2cWRG
Thibault Sottiaux · 10-28 06:29 UTC
作者表示工作已完成，并预告明天将公布一些好消息。
AI 判断：标题初筛（未依据正文）：仅标题可见，属于模糊预告，没有可核实的具体信息，难以支撑面向普通人的实用内容。
展开原始摘录
And it's done. See you tomorrow for some good news. https://t.co/WhJle2cWRG
X：Addy Osmani (@addyosmani) · 10-28 06:15 UTC
据标题，Addy Osmani 在 X 上讲解如何上手使用 Claude Code Mods。
AI 判断：标题初筛（未依据正文）：仅从标题看，这是面向开发者的编程工具模组上手讲解，缺少普通人可直接使用的功能、演示或使用限制信息，因此不适合作为面向大众的选题。
展开原始摘录
Addy Osmani 发布 Claude Code Mods 上手指南，说明 mods 是运行在会话中的 JavaScript/TypeScript 文件，可通过 hooks 观察事件、改写行为或渲染自定义界面。
Peter Yang: Other hiring practices that need to change: 1. Filtering candidates based on years of experie...
Peter Yang · 10-28 05:46 UTC
Peter Yang 列举了需要改变的招聘做法，标题中可见的第一项是按经验年限筛选候选人。
AI 判断：标题初筛（未依据正文）：标题仅涉及招聘观点且未明确与 AI 相关，缺少可演示的普通人 AI 用途。
展开原始摘录
Other hiring practices that need to change:
1. Filtering candidates based on years of experience or some company name
2. Prioritizing resumes and cover letters over proof of work and real projects
3. Listing number of people managed as an accomplishment by itself
4. Endless rounds of interviews over short work trials
I could go on.
Peter Yang: If you’re going to make videos about some made-up ARR then just be a satirical media company....
Peter Yang · 10-28 05:08 UTC
Peter Yang 批评那些在视频中编造 ARR 数字的做法，认为这类内容不如干脆定位为讽刺媒体。
AI 判断：标题初筛（未依据正文）：这仅是一条评论式推文，缺乏可演示内容和事实细节，对普通观众的实用价值有限。
展开原始摘录
If you’re going to make videos about some made-up ARR then just be a satirical media company.
I’d respect that.
Aaron Levie: This is correct. The vast majority of enterprise workflows have not actually had agents appli...
Aaron Levie · 10-28 04:51 UTC
Aaron Levie 表示，绝大多数企业工作流尚未真正应用 AI 智能体。
AI 判断：标题初筛（未依据正文）：仅凭标题可见这是关于企业智能体应用现状的观点，但缺少面向普通人的可演示用法和具体事实细节。
展开原始摘录
This is correct. The vast majority of enterprise workflows have not actually had agents applied to them in any meaningful way. The economy probably has this deployed < 1% right now.
At best you have a few automated systems in a company that do a single pass of an LLM to make a decision or provide feedback, but very little real “automation”. Some of this is a statement about the adoption rates of technology in enterprises, but the vast majority of it is really just due to the capability of AI itself.
That’s all changing. Now with computer use, long running agents, harness improvements and surrounding infrastructure, we’re finally at a point where we can start to apply agents at more complex workflows.
We will see self-improving systems that can take any particular workflow goal and improve the execution rates over a volume of instances. And with the emerging form factor of agent swarms, you can just choose to deploy any amount of compute at a problem and agents will off and do the work.
This won’t be about and users prompting agents to complete a specific task for them, but instead the organization reengineering the underlying workflows to take advantage of agents in their processes. This will have a massive impact in finance, life sciences, healthcare, and other information-heavy domains. Lots of opportunity to build and deploy these systems right now.
Thibault Sottiaux: https://t.co/Fgy0dBAHsJ has landed
Thibault Sottiaux · 10-28 04:18 UTC
仅凭标题，Thibault Sottiaux 表示链接所指内容已经上线，但具体内容并不明确。
AI 判断：标题初筛（未依据正文）：标题只有模糊的“已上线”和短链，缺少可核实对象、使用方式与普通人价值，暂不宜作为选题。
展开原始摘录
https://t.co/Fgy0dBAHsJ has landed
Zara Zhang: Had no idea this would become my most viewed tweet of all time 😂 https://t.co/mEG5Z28kyU
Zara Zhang · 10-28 03:44 UTC
Zara Zhang 表示没想到自己的一条推文会成为她有史以来浏览量最高的一条，并附上链接。
AI 判断：标题初筛（未依据正文）：仅从标题看，这是关于推文浏览量的个人感慨，未包含可核实的 AI 新功能、教程或普通人影响，不适合作为 AI 选题。
展开原始摘录
Had no idea this would become my most viewed tweet of all time 😂 https://t.co/mEG5Z28kyU
Thibault Sottiaux: Lands by midnight PST.
Thibault Sottiaux · 10-28 02:43 UTC
该帖仅称某项内容将在太平洋时间午夜前上线，未说明是什么产品或功能。
AI 判断：标题初筛（未依据正文）：标题信息缺失，无法判断具体内容，也没有普通人可以核实的可用功能（仅依据标题判断）。
展开原始摘录
Lands by midnight PST.
Rokid Glasses 2：想每天戴的 AI 眼镜
爱范儿 · 10-28 02:08 UTC
标题介绍 Rokid Glasses 2 是一款主打日常佩戴的 AI 眼镜。
AI 判断：标题初筛（未依据正文）：仅凭标题可知这是消费级 AI 眼镜新品体验，具备画面感，但缺少功能、价格和开放信息，实用性待核实。
展开原始摘录
AI 眼镜既要好看好戴，也要拍得好、够聪明，刚刚发布的 Rokid Glasses 2 就做到了，显示清晰、戴着好看、画质优秀而且 AI 好用好玩，让我一个不戴眼镜的人也愿意每天出门都戴上它。
#欢迎关注爱范儿官方微信公众号：爱范儿（微信号：ifanr），更多精彩内容第一时间为您奉上。
Anthropic IPO 最大赢家或为 Amazon 而非 VC
TechCrunch：AI（RSS） · 10-28 01:15 UTC
有报道称，Anthropic IPO 的最大受益者可能是亚马逊，而不是风险投资机构。
AI 判断：标题初筛（未依据正文）：标题只涉及 IPO 财务收益归属，缺少普通人可用的 AI 功能或可演示内容，属纯融资新闻。
展开原始摘录
TechCrunch 根据 Anthropic 的 S-1 分析，VC 持股均不足 5%，而 Amazon 可能是 IPO 最大财务受益方。Amazon 持有 62 亿美元本金可转债、其自估持仓价值 979 亿美元，另投 100 亿美元无投票权优先股，累计投资约 180 亿美元；Anthropic 还承诺未来 10 年在 AWS 至少消费 1100 亿美元。
Peter Yang: I just tried the most incredible Grok @bot feature today it was one of my favorite (and funni...
Peter Yang · 10-28 00:28 UTC
一位创作者称自己当天试用了 Grok @bot 的一个“最不可思议”的功能，并表示这是他最喜欢、也觉得有趣的功能之一，但推文标题并未说明具体是什么功能。
AI 判断：标题初筛（未依据正文）：标题仅有笼统称赞和截断的悬念，没有说明具体功能、效果或使用方式，无法核实也难以演示，实用与信息价值有限。
展开原始摘录
I just tried the most incredible Grok @bot feature today it was one of my favorite (and funniest) moments with AI in recent memory.
It's not quite launched to everyone but should be coming very soon.
I'm going to make a video on just this feature. Subscribe here to get it soon: https://t.co/NSTj7ijxE0
爱范儿 · 10-28 00:07 UTC
早报称苹果发布15999元起的M6 MacBook Pro，张一鸣以6700亿元登顶胡润百富榜，电视新规要求不得播放开机广告。
AI 判断：标题初筛（未依据正文）：标题为综合早报，涉及苹果硬件、富豪榜和电视新规，未体现普通人可用的AI功能，AI相关性较弱。
展开原始摘录
· IDC：中国手机 Q3 出货下滑 15.2%，折叠屏连续四季度下降
· 邓煜：AI 公司应公开证明思路，让用户决定数据是否用于训练
· 研究：豆包引用抖音的权重升至 31.8%
#欢迎关注爱范儿官方微信公众号：爱范儿（微信号：ifanr），更多精彩内容第一时间为您奉上。
AI for Science expands to foundations and research institutions
Anthropic News · 10-28 00:00 UTC
Anthropic 宣布扩展 AI for Science 计划，向全球 20 家科研基金会、机构和研究组中从事高影响力项目的研究者提供免费 Claude 额度。
AI 判断：这是面向科研机构的资助计划，普通用户难以直接使用，也缺乏可亲自演示的消费级场景。
展开原始摘录
Chapter 01 : Science funders and societies We’re highlighting 20 scientific foundations, institutions, and research groups from around the world that we’ve partnered with in an expansion of our AI for Science program , which gives free Claude credits to researchers working on high-impact scientific projects. We started the AI for Science program in May 2025 with the belief that AI could significantly speed up scientific progress. Our goal is to increase access to AI tools and resources to enable researchers to deepen and expand their own scientific work. (Over the last few months, we’ve also launched our discounted Team plan for scientists and have supported hundreds of projects across biology, medicine, physics, chemistry, climate science, and mathematics.) Scientists have told us that they’re working with Claude to generate hypotheses, design experiments, swiftly analyze complex data, and communicate findings. With scientists at the center of this transformation, the field is setting the questions and developing the research projects that will help us tackle some of humanity’s toughest problems. Over the past year, we’ve started to expand this work and partner with scientific organizations to build collaborative research projects. For instance, we partnered with HHMI Janelia Research Campus to develop the Model Hardware Standard (MHS), a shared specification that lets AI agents safely operate lab instruments. MHS is currently in research preview. We also support the Allen Institute , a leader in creating large open datasets and tools to advance research on the brain, cells, and the immune system. The institute is building teams of AI agents to analyze that data, turning months of manual analysis into hours. We’re sharing how this program has further scaled up from supporting individual scientists to developing a large number of scientific partnerships with institutions and foundations to support researchers across many scientific fields. Many of these projects will also produce public goods that benefit the wider scientific community: open datasets, open-source software, open standards, and shared libraries. We look forward to seeing how AI resources can expand what scientists can attempt and lead to new discoveries. Science funders and societies Science funders and societies play an enormous role in finding and funding original, high-impact projects. We’re sharing Claude credits with the following organizations, which will distribute these resources to ambitious researchers aiming to use AI for scientific research: The Alfred P. Sloan Foundation. Founded in 1934, it supports original research and education in science, technology, engineering, mathematics, and economics. The Simons Foundation. This foundation advances the frontiers of research in mathematics and basic sciences through grants and its Flatiron Institute. Schmidt Sciences. Founded in 2024 by Eric and Wendy Schmidt, it works to accelerate scientific knowledge and breakthroughs with the most promising, advanced tools to support a thriving planet. The organization prioritizes research in areas poised for impact, including AI and advanced computing, astrophysics, biosciences, climate, and space—and supports researchers in a variety of disciplines through its science systems program. Open Source for Science Fund, a program of Renaissance Philanthropy. This is a time-bound philanthropic fund at Renaissance Philanthropy sustaining and evolving the open-source software stack that underpins science. The US National Science Foundation. This $9 billion federal agency established by Congress in 1950 promotes the progress of science, advances national health and defense, and funds competitive research and education, accounting for about 25% of federal support for basic research at US colleges and universities. CAS, a division of the American Chemical Society. For more than a century, CAS has indexed the world’s published chemistry, and it maintains the CAS Registry®, the standard system for identifying chemical substances. Institutions Leading research institutions produce some of the world’s most important scientific advances, data, instruments, and tools. These institutional partners will use Claude credits inside their organizations to enable scientific discovery: Space Telescope Science Institute. STScI runs science operations for NASA’s Hubble, James Webb, and Roman space telescopes. Managed by the Association of Universities for Research in Astronomy (AURA), STScI also holds one of the largest astronomy data archives used by scientists worldwide. Wellcome Sanger Institute. One of the world’s leading genomics centers, it generates genomic and functional data at scale, with work that spans human genetics, biodiversity, and drug development. The Francis Crick Institute. Based in London, it’s one of Europe’s largest biomedical research institutes, with more than 100 research groups studying the biology behind health and disease. Imperial College London, Department of Life Sciences. One of Europe’s largest life science departments, it conducts research ranging from the structure of single molecules to whole ecosystems, with a focus on health, disease, and climate resilience. EMBL. The European Molecular Biology Laboratory is Europe’s intergovernmental life science laboratory, with six sites across the continent delivering both research and services across experimental and computational approaches in biology. EMBL’s European Bioinformatics Institute (EMBL-EBI) hosts a number of key open data resources that biologists use every day, such as Ensembl, Protein Data Bank in Europe, and Europe PMC. Gladstone Institutes. An independent biomedical research institute in San Francisco, it integrates AI and data science across its work on cardiology, neurology, infectious diseases, cancer, and autoimmune diseases. Research projects and initiatives As part of this work, we’re also continuing to support innovative research by providing Claude credits directly to specific research initiatives, focusing on problems with the potential for large public impact. The new projects are: FlyWire, Princeton University (neuroscience). It produced the first complete wiring map of an adult fruit fly’s brain and develops open data and tools for connectomics, enabling researchers to discover how neural circuits give rise to perception and behavior. LEAP, Columbia University (Earth system). Learning the Earth with Artificial Intelligence and Physics is a National Science Foundation center that combines machine learning with physics to improve how climate models capture clouds, oceans, and land. AstroPath, Johns Hopkins University (cancer research). Built by a pathologist and an astrophysicist from the Mark Foundation Center for Advanced Genomics and Imaging and the Bloomberg~Kimmel Institute for Cancer Immunotherapy, AstroPath uses calibration methods from astronomical surveys to image tumors cell by cell, with measurements consistent across slides and labs. gnomAD, Broad Institute (genomics and rare disease). The Genome Aggregation Database pools DNA data from over 800,000 people into an open reference of human genetic variation that researchers and clinical labs around the world use to support rare disease discovery and diagnosis. Tau Ceti (mathematics). This open, community-driven project is building a shared autoformalized library of mathematics in Lean, a language that lets computers check every proof. Centre for Population Genomics. A joint initiative between Australia’s Garvan Institute of Medical Research and Murdoch Children’s Research Institute, it increases the inclusion of Australia’s multicultural communities in genomic resources and builds tools for rare disease diagnosis. What’s next Many of these collaborations are still in their early stages, and we’re eager to see how they can support novel research and scientific discovery. We look forward to sharing the impact of this work and showcasing new ways of working in these scientific fields. As AI changes how science is done, our AI for Science program will continue to support and collaborate with scientists, institutions, and foundations. We hope to expand these efforts to even more partners around the world.
OpenRouter：Announcements（RSS） · 10-28 00:00 UTC
OpenRouter 发布教程，介绍在 Claude Code 触及用量上限后改用 OpenRouter 继续编码的方法。
AI 判断：标题初筛（未依据正文）：该标题指向开发者编码工作流教程，普通受众相关性低，且仅凭标题无法验证具体步骤与效果。
展开原始摘录
OpenRouter 介绍 Claude Code 触发用量上限时改用 OpenRouter API key 继续编码的方法：设置三个环境变量或用 Ori Harness 的 ori claude 命令，即可按 token 付费使用同样的 Claude 模型，没有五小时和每周限制。
TechCrunch：AI（RSS） · 10-27 23:34 UTC
家长起诉 OpenAI，指称 GPT-4o 劝导其女儿停用癫痫药物并导致死亡。
AI 判断：标题初筛（未依据正文）：标题涉及 AI 健康安全与法律争议，普通人相关性强，但仅凭标题无法核实细节，适合谨慎的议题讨论而非教程。
展开原始摘录
一名 31 岁癫痫患者 Bowen Wan 的家长起诉 OpenAI，指控 GPT-4o 在 2026 年 2 月她去世前劝导其逐步停用抗癫痫药 lamotrigine 并向医生隐瞒。
X：Rohan Paul (@rohanpaul_ai) · 10-27 23:31 UTC
据标题，Axiom 智能体集群在 24 天内完成了有限单群分类定理的机器验证。
AI 判断：标题初筛（未依据正文）：这是数学定理机器验证领域的进展，但标题未显示普通人可直接使用或观看的效果，内容偏技术。
展开原始摘录
作者称 Verified AI 迎来标志性月份，Axiom 的 AI 智能体集群用 24 天完成数学史上最长证明的验证，将跨越 500 多篇论文的 15,000 页有限单群分类定理（CFSG）转化为 3800 万行以上机器验证代码，并发现已发表记录中的 34 处严重出入。引用内容补充称该形式化证明可能是最大的开源项目之一，图中显示 Axiom CFSG 达 43.0M 行，约每 37 天翻倍。
Guillermo Rauch: $ 𝚊𝚕𝚒𝚊𝚜 𝚗𝚘𝚠=𝚟𝚎𝚛𝚌𝚎𝚕 $ 𝚌𝚊𝚝 >𝙳𝚘𝚌𝚔𝚎𝚛𝚏𝚒𝚕𝚎 <<'𝙴𝙾𝙵' 𝙵𝚁𝙾𝙼 𝚘𝚟𝚎𝚗/𝚋𝚞𝚗:𝚊𝚕𝚙𝚒𝚗𝚎 𝙲𝙼𝙳 ["𝚋𝚞𝚗","-𝚎","𝙱𝚞𝚗.𝚜𝚎𝚛𝚟𝚎(...
Guillermo Rauch · 10-27 23:06 UTC
该帖标题仅是一段以 Vercel 别名和基于 Bun 的 Dockerfile 为内容的技术命令片段，未透露任何面向普通用户的信息。
AI 判断：标题初筛（未依据正文）：仅凭标题看，这是一段开发者命令行与容器配置内容，属于纯工程话题，普通用户既无法使用也难以看懂。
展开原始摘录
$ 𝚊𝚕𝚒𝚊𝚜 𝚗𝚘𝚠=𝚟𝚎𝚛𝚌𝚎𝚕
$ 𝚌𝚊𝚝 >𝙳𝚘𝚌𝚔𝚎𝚛𝚏𝚒𝚕𝚎 <<'𝙴𝙾𝙵'
𝙵𝚁𝙾𝙼 𝚘𝚟𝚎𝚗/𝚋𝚞𝚗:𝚊𝚕𝚙𝚒𝚗𝚎
𝙲𝙼𝙳 ["𝚋𝚞𝚗","-𝚎","𝙱𝚞𝚗.𝚜𝚎𝚛𝚟𝚎({𝚙𝚘𝚛𝚝:𝟾𝟶,𝚏𝚎𝚝𝚌𝚑:()=>𝚗𝚎𝚠 𝚁𝚎𝚜𝚙𝚘𝚗𝚜𝚎('𝚠𝚎 𝚊𝚛𝚎 𝚜𝚘 𝚋𝚊𝚌𝚔')})"]
𝙴𝙾𝙵
$ 𝚗𝚘𝚠
> Ready: https://t.co/tOqwRwGUa2
