Meta发布Muse个人AI助手,基于Spark 1.3模型
Meta just released Muse AI assistant for personal tasks, powered by Muse Spark 1…
推荐理由
Meta正式推出具备长程执行能力的个人Agent产品Muse,其VM隔离与凭证分离的安全设计对行业有参考价值,建议关注其实际落地效果。
Meta just released Muse AI assistant for personal tasks, powered by Muse Spark 1.3, Meta's latest model for longer-horizon agentic work
Meta 刚刚发布了面向个人任务的 Muse AI 助手,该助手由 Meta 最新推出的、适用于长周期智能体工作的 Muse Spark 1.3 模型驱动
- it can connect to email, calendars, shopping, payments and other services, then keep working after the app closes.
- each user gets a Muse Secure VM, with a separate Sentinel agent that checks outbound data and actions before they reach the network.
- Meta is making Muse free to use for up to 100M tokens per week, with subscription plans for people who use more compute
- Sensitive actions such as purchases or emails require user approval, while credentials stay in Secure Credential Storage so Muse cannot read passwords or payment details.
- the model, Muse harness, deterministic code and classifiers screen prompt injections before hostile content can enter the model's context.
- currently it launches for U.S. users on web, iOS, Android and WhatsApp, with free access plus $20 and $100 monthly tiers for heavier usage.
- the security design moves the trust boundary beyond model behavior into VM isolation, credential separation and a kernel-enforced gate on network actions.
- 它可以连接电子邮件、日历、购物、支付等服务,并在应用关闭后继续执行任务。
- 每位用户都会获得一个 Muse 安全虚拟机(Secure VM),其中包含一个独立的 Sentinel 代理,用于在数据或操作到达网络之前进行检查。
- Meta 将 Muse 设为免费使用,每周提供高达 1 亿 token 的额度,对于需要更多计算资源的用户则提供订阅计划
- 敏感操作(如购买或发送邮件)需要用户批准,而凭据保存在安全凭据存储中,因此 Muse 无法读取密码或支付详情。
- 通过模型、Muse Harness、确定性代码和分类器,可在恶意内容进入模型上下文之前拦截提示注入攻击。
- 目前该服务面向美国用户在网页端、iOS、Android 和 WhatsApp 上线,提供免费访问权限,并设有每月 20 美元和 100 美元的付费层级以满足更高用量需求。
- 其安全设计将信任边界从模型行为扩展至虚拟机隔离、凭据分离以及对网络操作的基于内核的执行门禁控制。
更进一步:量化金融体系
看懂新闻只是起点——沿量化金融路径,把它变成能交付的工程能力