Pluto Security Raises $42M to Tackle the Risks of Employee-Built AI Tools

Employees building their own AI tools is becoming a new security blind spot. Pluto Security has closed $42 million in total funding to focus specifically on this area that traditional security systems have overlooked.

The Risk Isn't in Developers — It's in Finance and Legal

Shahar Bahat, co-founder and CEO of Pluto Security, shared a counterintuitive data point: about 60% of the risks the company detects come from non-developers.

The real source of the problem is the finance department. According to Bahat, finance teams rank first when it comes to "building tools without approval," followed by marketing and legal, while professional developers come in last.

This breaks a common assumption — most people assume security risks are primarily introduced by technical teams. The reality is that low-code and generative AI now allow non-technical staff to quickly build applications, and these applications often lack basic security review.

Pluto's tool once identified a typical case: a finance employee at a large company created a customer-facing application that was directly exposed to the public internet with no access controls.

What's meant by "building tools without approval" here isn't just Excel macros or scripts — it refers to employees using low-code platforms, no-code platforms, and generative AI code generation capabilities to assemble applications, automated workflows, or data interfaces that can serve external users. These tools typically bypass IT department procurement and security reviews, without threat modeling, penetration testing, or permission hardening. Once shared beyond the team, they can directly expose corporate data or internal systems to the outside. The widespread adoption of low-code has given finance, legal, marketing, and other business roles "developer" capabilities — this is the fundamental reason the source of risk is shifting from technical teams to business departments.

The Spread of AI Tools Outpaces Security Teams' Ability to Respond

Bahat noted that new AI tools can be deployed within an enterprise within two hours of release. This speed means traditional security approval processes can barely keep up.

An employee discovers a useful AI tool, tries it, integrates it, shares it with colleagues — the whole process may be complete before the security team has even heard the tool's name.

Pluto's software targets exactly this scenario: continuously monitoring the AI tools employees use and the applications they create, identifying potential risks. The company has also launched a free platform called Plutonium that allows risk assessment of public AI tools.

Traditional enterprise security governance typically relies on procurement approval, access assessment, and pre-launch security testing, with a tool introduction cycle often measured in weeks or even months. But generative AI tools largely exist as browser extensions, SaaS trial accounts, API keys, and the like — employees can self-register and start using them immediately without IT department involvement. So enterprises no longer face the question of "whether to allow a certain tool" but rather "how many tools are already in use, and which ones are handling company data." Visibility therefore becomes a prerequisite for governance: only by first discovering shadow AI can you assess its data flows, permission scope, and compliance risks, then decide whether to keep, restrict, or replace it.

Business Progress: 30+ Customers, $2M ARR

Pluto Security began selling its software last April and has already surpassed $2 million in annual recurring revenue (ARR).

The company has more than 30 enterprise customers, including Oscar Health, Workato, Sixt, and AlphaSense. It currently has 40 employees across the US and Israel.

On the funding side, the company first closed an $8.5 million seed round led by TLV Partners, followed by a $34 million follow-on round led by Jump Capital, with participation from investors including Skinos and Ansa Capital. The two rounds total $42 million.

ARR (Annual Recurring Revenue) is a core metric for measuring the revenue quality of SaaS and subscription-based businesses, typically calculated by multiplying current monthly recurring revenue by 12. It reflects the scale of repeatable, predictable revenue rather than one-time contracts or service income. For Pluto Security, which is still in its early stages with enterprise subscriptions as its primary business model, surpassing $2 million in ARR means the product has completed the leap from proof of concept to paid renewals. This figure should also be considered alongside customer count: 30-plus customers corresponding to $2 million ARR suggests that per-customer deal sizes fall within the typical range for mid-to-large enterprise security software, with a customer base primarily composed of mid-to-large enterprises.

Why This Space Is Worth Watching Now

Pluto Security's funding reflects a broader trend: the path of AI penetration within enterprises is shifting from "unified procurement by IT departments" to "spontaneous employee adoption."

When the AI tool adoption curve is far steeper than the security governance curve, the gap in between becomes the attack surface — applications exposed to the public internet, unaudited data flows, API calls lacking permission management.

Pluto has chosen to enter at the "visibility" layer: first let enterprises know which AI tools their employees are using and what applications they've built, then discuss controls. This approach is similar to the logic CASB (Cloud Access Security Broker) used to solve the shadow IT problem years ago — except the target has shifted from SaaS to AI tools and employee-built applications.

As generative AI continues to lower the barrier to building applications, this kind of "non-developer risk" will only increase, not decrease. The $42 million injection shows investors agree with this assessment.

CASB (Cloud Access Security Broker) is a mature category in cloud security, deployed between users and cloud services to discover shadow IT, enforce access controls, encrypt sensitive data, and meet compliance requirements. The core problem it solves is: after employees bypass IT departments to use SaaS on their own, enterprises lose control over data flows. What Pluto does is logically akin to CASB, but its targets extend from traditional SaaS to AI tools and employee-built applications, making identification more difficult — because many AI tools lack standardized enterprise interfaces, and employee-built applications may run on personal accounts or public clouds, making traditional network-boundary-based detection methods hard to apply.

One-line Summary

Pluto Security raises $42M to target the security blind spot of employee-built AI tools.

Paragraph Summary

Pluto Security has closed $42 million in total funding, with an $8.5 million seed round led by TLV Partners and a $34 million follow-on round led by Jump Capital. The company targets the security blind spot created by employees building AI tools without approval. Its monitoring data shows about 60% of risks come from non-developers, with finance ranking first, followed by marketing and legal, while professional developers come last. AI tools can be deployed within an enterprise within two hours of release, making traditional security approval processes hard to keep up. Pluto identifies risks by continuously monitoring the AI tools employees use and the applications they build. It has surpassed $2 million in ARR with 30-plus enterprise customers including Oscar Health, Workato, Sixt, and AlphaSense, and has launched a free platform called Plutonium to assess the risks of public AI tools.