Pluto Security has raised $42 million as enterprises scramble to manage a new kind of risk: employees rapidly adopting AI tools that can create exposed apps, risky connectors and other security gaps before IT teams fully understand them. The startup says the problem is moving so fast that newly released AI tools can appear inside a company’s environment within hours, and that security teams need visibility into not just users, but the behavior of AI agents themselves.
Pluto Security’s pitch: secure the agent, not just the user
Pluto Security CEO and co-founder Shahar Bahat argues that traditional enterprise security tooling is not built for the current wave of AI adoption. In her view, companies are moving from monitoring user behavior to monitoring agent behavior, because employees are increasingly building and deploying AI-powered workflows that operate semi-independently.
The startup’s software is designed to track the AI tools employees use, along with the “skills,” extensions, MCP servers and related components those tools depend on. It also looks at the apps and agents being created, aiming to flag risky components and suspicious activity without forcing organizations to block AI use entirely.
What Pluto says it is seeing inside enterprises
Bahat told TechCrunch that roughly 60% of the risks Pluto detects across its customer base come from tools vibe-coded by non-developers. Finance teams are the largest source of those risks, followed by marketing and legal, with developers ranking fourth, she said.
She pointed to one example the company found at a Fortune 500 company: a finance employee had built a vibe-coded app to manage customer information and churn, but it was left exposed to the internet. Pluto says this kind of issue is exactly why AI security needs to focus on what employees are creating, not only what software they are formally approved to use.
Live threats during proof-of-concepts
According to Bahat, the company often finds active threats while testing with potential customers. “In general, I’ll say there was not one POC in which we entered and didn’t find malicious stuff,” she said. “Whether those are extension packages or skills — some of them had active live attacks while we did the POC. Which again, for me, product-wise is amazing. For the companies, it’s not very good.”
That’s a reminder that the security risks around AI tools are not just theoretical. As organizations rush to adopt generative AI and agentic workflows, they may also be introducing new attack surfaces faster than their existing controls can adapt.
How Pluto Security’s system works
Pluto says its platform connects to endpoint security and device management systems, or installs its own sensors directly on devices. It gathers information such as prompts, commands, running processes, permissions and a user’s previous activity to build context around what an agent is doing.
The company also says it uses its own “lightweight” models on endpoints to assess the agent’s behavior in real time. That allows the system to intervene mid-session if necessary, rather than simply logging activity after the fact.
- Monitors AI tools employees use
- Tracks related components such as skills, extensions and MCP servers
- Detects risky components and dangerous agent activity
- Can enforce policies without stopping AI adoption
- Can stop an action or inject context into a session when behavior looks unsafe
One example Bahat gave: if an agent tries to gain permissions beyond what a user intended, Pluto can stop the action or add instructions into the session that prompt the agent to ask for confirmation. “We can inject context back to the session, and this way I can change how the agent is working,” she explained.
A fast-growing market for AI security tools
Pluto is part of a broader crop of startups built around the idea that enterprises need dedicated controls for AI tools, connectors and agent workflows. The company’s argument is that existing security software is a poor fit for this new environment, making the category a natural target for fresh investment.
The startup says it began selling in April and has already crossed $2 million in annual recurring revenue. It also says it has more than 30 enterprise customers, including Oscar Health, Workato, Sixt and AlphaSense.
The funding behind Pluto’s launch
Pluto’s $42 million in funding includes an $8.5 million seed round led by TLV Partners, with cybersecurity firm Armis’ founders also investing. The company later raised a $34 million Series A led by Jump Capital, with participation from Skinos, the venture firm of Check Point co-founder Shlomo Kramer, Ansa Capital and other angel investors.
The funding will go toward product development, expanded security research and a larger U.S. presence. Pluto said it has 40 employees across the U.S. and Israel.
Why Pluto thinks demand will keep rising
Bahat says the risks are not limited to the biggest AI-heavy technology companies. She described the shape of the problem as dependent on the CISO’s mindset and the organization’s appetite for AI adoption, but said threats show up across environments.
That helps explain why Pluto’s Plutonium platform has become an important funnel for the company. Launched in April, it lets anyone check the risks associated with publicly available AI skills, connectors and add-ons. It also includes an MCP scanner and a catalog of verified components.
Bahat said thousands of security professionals use Plutonium, and that 70% of Pluto’s sales pipeline now comes from inbound interest, though she did not specify how much of that comes directly from the free platform.
Source: Original report
Was this helpful?
Explore more: Application Audit & Review More Business & Startups Tech News
Last Modified: October 13, 2026 at 10:31 pm
0 views
