供资
Sentra 融资 150M 为 AI Agents 保护企业数据
在 Google 上的首选来源中添加 Unite.AI
森特拉获得了由Capital One Ventures牵头的1.5亿美元战略资本投资，以扩大其数据安全平台，因为企业让AI agents获得更多的商业信息。 宣布 11月4日，投资带来资本一号风险投资公司与现有投资者贝塞默风险投资伙伴，采耶夫风险投资公司，标准投资公司，慕尼黑再风险投资公司和Key1资本公司并列。
一名代表将加入森特拉的董事会。 称公司的投资年增长率超过300%，尽管公告没有具体说明该数字背后的财务措施或披露估值。
随着人工智能系统从回答问题转向执行工作，筹资目标更加突出：agents可以继承获取一个组织从未完全分类或管理过的敏感信息的渠道。
保护数据 Agents Can Reach
Sentra的做法始于云层存储、SaaS应用程序、仓库、档案共享和AI使用的知识库。 它发现敏感信息，将其分类，并绘制能够访问的人，服务账户和agents的地图，包括通过继承的许可。
这与agent行动时检查每一个即时或工具呼叫不同。 拥有不必要宽泛权限的数据集，即使其用户从未打算将其曝光，也可以被助手访问。 提前减少这种暴露可缩小错误指示或身份受损的后果。
“AI agents现在是企业数据最繁忙的消费者，agents以机器速度移动，各组织能够做的是确保agents所达到的数据已经安全。”
发现、获取治理和应对
森特拉描述了三个相连的阶段。 在访问之前，它识别敏感记录，绘制地图权限，并帮助删除冗余，过时或高风险的数据。 在可疑活动期间，了解数据的检测可以标出异常访问和触发补救，包括取消访问或消除风险共享。 之后，调查人员可以检查一个身份可能到达的敏感数据集。
该平台结合了数据安全态势管理(DSPM)，数据检测和响应，数据访问治理和数据损失预防。 Sentra的DSPM指南提供了发现和暴露管理层的背景：了解数据存放地点、如何使其敏感以及安全态势如何产生风险。
将分类与身份事项联系起来，因为两个观点都不够。 高度敏感的文件可能会受到适当的限制，而广泛共享的文件夹可能没有产生什么后果。 当一个小组能够评估信息，并有权共同获取信息时，确定优先顺序就更好了。
分析其存在的信息
Sentra 称在客户环境中分析客户数据，而不是将敏感记录转移到单独的由供应商控制的数据产业区。 其公共平台材料描述持续发现，分类，数据卫生和接入治理是AI准备的基础。
对于大型或受监管的企业，这种架构可以减少获得另一份敏感信息的必要性。 它并不消除配置权限、建立覆盖范围和维持补救工作流程的必要性。 该公告关于规模和成本优势的声称是公司报表，而不是独立披露的比较结果。
实际目标是在agent到达之前使数据环境更加安全，然后保留足够的上下文来控制和调查出乎意料的行为。 这补充了agent执行周围的控制，而不是取代所有其他的AI安全层。
产品扩展和市场拓展资本
Sentra计划投资对agents,copilots和模型的能见度，持续对人和非人身份进行最少的准入，并在agent行为出乎意料时更快的遏制。 它还将扩大现场和频道节目、营销及其地理覆盖范围。
Capital One Ventures自称是数据、技术和金融服务公司的战略投资者。 它的参与和董事会代表将融资与投资者联系起来，直接关系到企业如何管理宝贵信息。 发行版未注明与首都一号单独签订商业部署协议。
森特拉成立于2021年，总部设在纽约和特拉维夫，由约夫·雷格夫，阿萨夫·科昌，罗恩·雷特尔和亚伊尔·科恩建立。 公司领导网页分别列出他们为CEO,CTO和CPO总裁。
投资支持一个以AI系统下的信息为中心的安全论文。 随着更自主的软件获得企业资源，了解存在哪些数据，以及谁或谁能够接触到，成为控制由此而来的风险的一个先决条件。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Funding
Sentra Raises $150M to Secure Enterprise Data for AI Agents
Add Unite.AI to your preferred sources on Google
Sentra has secured a $150 million strategic capital investment led by Capital One Ventures to expand its data-security platform as enterprises give AI agents access to more business information. Announced November 4, the investment brings Capital One Ventures alongside existing investors Bessemer Venture Partners, Zeev Ventures, Standard Investments, Munich Re Ventures and Key1 Capital.
A Capital One representative will join Sentra’s board. The company says the investment follows more than 300% year-over-year growth, although the announcement does not specify the financial measure behind that figure or disclose a valuation.
The funding targets a problem that becomes more consequential as AI systems move from answering questions to executing work: agents can inherit access to sensitive information that an organization has never fully classified or governed.
Protecting the Data Agents Can Reach
Sentra’s approach begins with data at rest across cloud storage, SaaS applications, warehouses, on-premises file shares and knowledge stores used by AI. It discovers sensitive information, classifies it and maps the people, service accounts and agents able to access it, including through inherited permissions.
That is a different control point from examining every prompt or tool call as an agent acts. A dataset with unnecessarily broad permissions can become accessible to an assistant even when its user never intended to expose it. Reducing that exposure in advance can narrow the consequences of a mistaken instruction or compromised identity.
“AI agents are now the busiest consumers of enterprise data, and many of them inherit access to data that hasn’t been fully classified or governed,” said Yoav Regev, Sentra’s co-founder and CEO, in the announcement. “Traditional perimeter controls weren’t built for agents moving at machine speed. What organizations can do is make sure the data those agents reach is already secured.”
Discovery, Access Governance and Response
Sentra describes three connected stages. Before access occurs, it identifies sensitive records, maps permissions and helps remove redundant, obsolete or high-risk data. During suspicious activity, data-aware detection can flag unusual access and trigger remediation, including revoking access or removing risky sharing. Afterward, investigators can examine the sensitive datasets an identity could have reached.
The platform combines data security posture management (DSPM), data detection and response, data access governance and data loss prevention. Sentra’s DSPM guide provides background on the discovery and exposure-management layer: understanding where data resides, what makes it sensitive and how its security posture creates risk.
Connecting classification with identity matters because neither view is sufficient alone. A highly sensitive file may be appropriately restricted, while a broadly shared folder may contain little of consequence. Prioritization improves when a team can evaluate the information and the authority to reach it together.
Analyzing Information Where It Lives
Sentra says it analyzes customer data within the customer’s environment rather than transferring the sensitive records into a separate vendor-controlled data estate. Its public platform materials describe continuous discovery, classification, data hygiene and access governance as the foundation for AI readiness.
For large or regulated enterprises, that architecture can reduce the need to secure another copy of sensitive information. It does not remove the need to configure permissions, establish coverage and maintain remediation workflows. The announcement’s claims about scale and cost advantages are company statements rather than results of an independently disclosed comparison.
The practical objective is to make the data environment safer before an agent reaches it, then retain enough context to contain and investigate unexpected behavior. This complements controls around agent execution rather than replacing every other layer of AI security.
Capital for Product Expansion and Market Reach
Sentra plans to invest in visibility into agents, copilots and models, continuous least-privilege access for human and non-human identities, and faster containment when an agent behaves unexpectedly. It will also expand field and channel programs, marketing and its geographic reach.
Capital One Ventures describes itself as a strategic investor in data, technology and financial-services companies. Its participation and board representation connect the financing to an investor with a direct interest in how enterprises govern valuable information. The release does not identify a separate commercial deployment agreement with Capital One.
Founded in 2021 and headquartered in New York and Tel Aviv, Sentra was established by Yoav Regev, Asaf Kochan, Ron Reiter and Yair Cohen. The company’s leadership page lists them as CEO, president, CTO and CPO, respectively.
The investment backs a security thesis centered on the information beneath AI systems. As more autonomous software gains access to enterprise resources, knowing what data exists and who or what can reach it becomes a prerequisite for controlling the resulting exposure.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.