What are the latest funding news in the AI safety market? (October 2026)
Whether you're building or investing in the AI safety market, our report covers everything you need to know: the players, the money, and what's coming next.
AI safety funding accelerated sharply in September 2026, with investors backing companies that can control, test, monitor and secure increasingly autonomous AI systems.
The latest deals show a clear shift toward runtime AI agent security, where companies control what agents can access and what actions they can take inside real enterprise systems.
The market also includes a growing assurance layer around AI evaluation, red teaming, governance, hallucination prevention and independent certification.
And if you want to better understand this new industry, you can download our pitch covering the AI safety market.
Insights
- The 12 latest AI safety companies in this dataset raised about $449.9 million, with Alice, HiddenLayer and AIR alone representing $290 million, or roughly 64% of disclosed capital.
- Seven of the 12 companies directly address AI agents at runtime, showing how agent authorization, identity, monitoring and tool security are becoming central parts of the AI safety market.
- Runtime AI safety is moving closer to the point of execution: Kontext, Outerlimit and Eve Security all aim to inspect or control agent actions before sensitive enterprise systems are affected.
- AI safety investors are funding both sides of deployment risk: companies such as AIUC and Arga test systems before production, while HiddenLayer, Lasso and AIR protect them during live operation.
- The funding range is unusually wide, from Kontext's $4 million round to Alice's $140 million financing, showing that AI safety now supports both new infrastructure startups and scaled commercial platforms.
- September dominated the latest AI safety funding activity, with 10 of these 12 deals announced during the month and several rounds appearing within the first half of September alone.
- AI agent security is splitting into specialized product categories, including runtime authorization, identity governance, MCP and skill security, behavioral monitoring, adversarial testing and application guardrails.
- Independent AI assurance is emerging alongside technical security: AIUC combines audits, standards and insurance, suggesting that AI safety infrastructure may increasingly connect technical evaluations with financial risk transfer.
Written by the New Market Pitch teamWe track new markets so founders and investors can move faster. About us
In our report covering the AI safety market, you will always find fresh funding data, taken straight from our own verified funding database.
Summary table of the latest funding deals in the AI safety market as of October 2026
We define the AI safety market as products and services that reduce harms and failure modes from AI systems by measuring, mitigating, and governing model behavior and AI-specific risk.
We include AI evaluation and red teaming, safety monitoring and incident response, safety and guardrail layers, and AI trust, risk and security management capabilities such as governance, robustness, privacy and security controls, and explainability when used for risk reduction.
We exclude generic MLOps, general cybersecurity, and general compliance or content moderation offerings unless they are specifically designed to address AI-model behavior or AI-specific threats.
| Name | When | Amount in $ | Round Type | Category |
|---|---|---|---|---|
| Alice | 25 August 2026 | $140M | Growth / reported Series C | AI adversarial evaluation & safety guardrails |
| Arga Labs | 26 August 2026 | $10M | Seed | AI agent evaluation & simulation |
| AIR | 1 September 2026 | $50M total | Two seed rounds | AI agent supply-chain & MCP security |
| Lasso Security | 2 September 2026 | $30M | Series A | AI guardrails & runtime security |
| HiddenLayer | 2 September 2026 | $100M | Series B | AI model & agent security |
| AI Score | 3 September 2026 | $5.4M | Seed | AI governance & TRiSM |
| Resect AI | 3 September 2026 | $25M | Undisclosed financing | Model robustness & hallucination prevention |
| Cymphony | 9 September 2026 | $25M | Series A | AI agent identity & access governance |
| AIUC | 15 September 2026 | $40M | Series A | AI evaluation, assurance & risk transfer |
| Eve Security | 15 September 2026 | $4.5M | Seed extension | AI agent runtime monitoring & response |
| Outerlimit | 22 September 2026 | $16M | Pre-seed | AI agent authorization & runtime security |
| Kontext | 24 September 2026 | $4M | Seed | AI agent runtime authorization & guardrails |
All the latest funding deals in the AI safety market as of October 2026
Kontext raised $4M in September 2026 to expand its runtime security platform for AI agents.
When was it?
Kontext announced the financing on September 24, 2026.
Who are they?
Kontext provides a runtime security layer that checks AI agent actions against identity, context and security policies before allowing them to execute.
Geographical focus?
Kontext is based in Munich and targets organizations deploying workplace and coding agents across international markets.
Why do we include them in the AI safety market?
Kontext directly reduces unsafe AI agent behavior by controlling actions at runtime, placing it in the AI agent authorization and guardrails category.
What is the company stage?
Kontext is at the seed and early-commercial stage, with a working platform and initial customer deployments.
How much did they raise?
Kontext raised $4 million in this financing.
What round is it?
Kontext described the financing as a seed round.
Why did they raise?
Kontext raised the capital to expand engineering, develop its runtime enforcement platform and support more enterprise deployments.
In our research about AI safety, we turn what is vague and complex into simple, useful visuals. Market maps like this one show you at a glance where every startup stands.
Outerlimit raised $16M in September 2026 to build a security and authorization layer for autonomous AI agents.
When was it?
Outerlimit announced the financing on September 22, 2026, alongside its emergence from stealth.
Who are they?
Outerlimit controls AI agent actions by connecting identity, authorization and tool execution through a dedicated enterprise security layer.
Geographical focus?
Outerlimit operates from London and New York and targets large enterprises deploying AI agents internationally.
Why do we include them in the AI safety market?
Outerlimit is built specifically to prevent unauthorized or unsafe autonomous actions, placing it in AI agent runtime authorization and deterministic safety controls.
What is the company stage?
Outerlimit is at the pre-seed and post-stealth stage, with its initial security architecture moving into enterprise deployments.
How much did they raise?
Outerlimit raised $16 million.
What round is it?
Outerlimit raised a pre-seed round.
Why did they raise?
Outerlimit raised the funding to build its agent security architecture, grow the team and bring deterministic authorization controls into production environments.
AIUC raised $40M in September 2026 to expand independent audits, standards and insurance for advanced AI systems.
When was it?
AIUC announced the Series A on September 15, 2026.
Who are they?
AIUC audits and stress-tests AI systems against risks such as hallucinations, jailbreaks and data leaks, then connects certification with insurance coverage.
Geographical focus?
AIUC is based in San Francisco and targets enterprise AI developers and users internationally.
Why do we include them in the AI safety market?
AIUC directly measures and reduces AI-specific failure risks through testing and assurance, placing it in AI evaluation, auditing and risk-transfer infrastructure.
What is the company stage?
AIUC is at the early-commercial and PMF-building stage, with active certification infrastructure and enterprise customers.
How much did they raise?
AIUC raised $40 million in the round, bringing disclosed funding to $55 million.
What round is it?
AIUC raised a Series A.
Why did they raise?
AIUC raised the capital to expand AI audits, standards and insurance from enterprise agents toward more advanced frontier AI systems.
Eve Security raised $4.5M in September 2026 to strengthen real-time protection against dangerous AI agent behavior.
When was it?
Eve Security announced the additional funding on September 15, 2026.
Who are they?
Eve Security monitors autonomous AI agents while they operate and lets security teams detect and stop dangerous actions before sensitive systems are affected.
Geographical focus?
Eve Security is based in Austin and focuses on enterprise customers deploying autonomous AI systems.
Why do we include them in the AI safety market?
Eve Security exists specifically to identify and contain unsafe AI agent behavior, placing it in runtime monitoring, detection and response.
What is the company stage?
Eve Security is at the seed and early-commercial stage, following product validation and growing enterprise demand.
How much did they raise?
Eve Security raised $4.5 million in new capital, extending its total seed financing to $7.5 million.
What round is it?
Eve Security raised a seed extension.
Why did they raise?
Eve Security raised the capital to accelerate product development and commercial expansion as enterprises deploy more autonomous agents.
In our market report about AI safety, we always try to gather several sources and make estimates that make sense, so you can work with solid numbers.
Cymphony raised $25M in September 2026 as enterprises faced growing identity and data risks from AI agents.
When was it?
Cymphony announced the Series A on September 9, 2026.
Who are they?
Cymphony maps employees, AI agents, permissions, sensitive data and activity into a single workforce graph that helps security teams identify risky access.
Geographical focus?
Cymphony operates from New York and Tel Aviv, with most customers in North America and growing demand in Europe, the Middle East and Africa.
Why do we include them in the AI safety market?
Cymphony helps enterprises control AI agents as autonomous digital identities, placing it in AI agent identity, permissions and data-access governance.
What is the company stage?
Cymphony is at the early-PMF stage, with double-digit enterprise customers and seven-figure annual recurring revenue within its first year of sales.
How much did they raise?
Cymphony raised $25 million in the Series A.
What round is it?
Cymphony raised a Series A.
Why did they raise?
Cymphony raised the funding to scale its platform and expand across large enterprises adopting autonomous AI identities.
AI Score raised $5.4M in September 2026 to scale continuous governance for generative and agentic AI.
When was it?
AI Score announced the funding on September 3, 2026.
Who are they?
AI Score inventories enterprise AI systems, connects them to owners and risks, and continuously applies governance controls to models, applications and agents.
Geographical focus?
AI Score is based in London, with early traction in regulated UK organizations and plans to serve customers internationally.
Why do we include them in the AI safety market?
AI Score continuously manages AI-specific operational risk and controls, placing it in AI trust, risk and security management and AI governance.
What is the company stage?
AI Score is at the seed and early-PMF stage, with commercial adoption and rapid revenue growth.
How much did they raise?
AI Score raised $5.4 million.
What round is it?
AI Score raised a seed round.
Why did they raise?
AI Score raised the capital to accelerate platform development, expand go-to-market activity and deepen controls for autonomous AI agents.
Resect AI raised $25M in September 2026 to commercialize technology designed to detect hallucinations before they reach users.
When was it?
Resect AI announced the financing on September 3, 2026, when the company launched publicly from stealth.
Who are they?
Resect AI observes internal large-language-model behavior during inference and can detect, interpret and modify problematic behavior before a hallucination reaches the user.
Geographical focus?
Resect AI is rooted in the Pacific Northwest and targets enterprise AI deployments across the U.S. and broader markets.
Why do we include them in the AI safety market?
Resect AI directly addresses model reliability failures during inference, placing it in model robustness, hallucination prevention and AI accountability.
What is the company stage?
Resect AI is at the post-stealth and early-commercialization stage.
How much did they raise?
Resect AI raised $25 million.
What round is it?
Resect AI disclosed $25 million from private-equity investors without assigning the financing a conventional venture round label.
Why did they raise?
Resect AI raised the funding for research and development, go-to-market expansion and hiring in the Seattle and Portland regions.
Our document detailing everything you need to know about AI safety gets updated several times a month, so you always get the latest view of where things stand.
Lasso Security raised $30M in September 2026 to expand AI guardrails and runtime protection across enterprise deployments.
When was it?
Lasso Security announced the financing on September 2, 2026.
Who are they?
Lasso Security combines automated AI red teaming with runtime guardrails that inspect AI requests and block attacks or unsafe behavior.
Geographical focus?
Lasso Security operates from Tel Aviv and New York and is expanding across North America, Europe and regulated U.S. markets.
Why do we include them in the AI safety market?
Lasso Security directly protects models, AI applications and agents from AI-specific threats, placing it in guardrails, red teaming and runtime AI security.
What is the company stage?
Lasso Security is at the early-growth stage, with production deployments across major enterprises and U.S. government organizations.
How much did they raise?
Lasso Security raised $30 million.
What round is it?
Lasso Security's financing has been reported as a Series A.
Why did they raise?
Lasso Security raised the capital to expand engineering, deepen work with federal and regulated customers and grow sales across North America and Europe.
HiddenLayer raised $100M in September 2026 as enterprise demand for AI model and agent security accelerated.
When was it?
HiddenLayer announced the Series B on September 2, 2026.
Who are they?
HiddenLayer protects machine-learning models, generative AI applications and autonomous agents from adversarial attacks, malicious code and unsafe runtime behavior.
Geographical focus?
HiddenLayer is based in Austin and serves enterprise, technology, financial-services and U.S. government customers.
Why do we include them in the AI safety market?
HiddenLayer's core business is AI-specific security and risk reduction, placing it in model security, adversarial testing and agent runtime protection.
What is the company stage?
HiddenLayer is at the growth stage, with established product-market fit and annual recurring revenue in the tens of millions of dollars.
How much did they raise?
HiddenLayer raised $100 million.
What round is it?
HiddenLayer raised a Series B.
Why did they raise?
HiddenLayer raised the capital to deepen its enterprise security platform and expand runtime protection for autonomous agents and AI coding systems.
AIR revealed $50M in September 2026 to secure the tools, skills and software components used by enterprise AI agents.
When was it?
AIR publicly revealed both seed financings on September 1, 2026.
Who are they?
AIR discovers enterprise AI agents, continuously vets their skills, plug-ins and MCP servers, and blocks connections to components that fail security checks.
Geographical focus?
AIR has Israeli roots and is expanding commercially across the U.S. and Europe, particularly among regulated enterprises.
Why do we include them in the AI safety market?
AIR reduces AI-specific supply-chain and runtime risks around autonomous agents, placing it in AI agent security, MCP security and tool governance.
What is the company stage?
AIR is at the seed and early-commercial stage, emerging from stealth with more than 20 customers.
How much did they raise?
AIR disclosed $50 million across a $10 million initial seed and a subsequent $40 million seed financing.
What round is it?
AIR disclosed two seed rounds together rather than one single $50 million round.
Why did they raise?
AIR raised the capital to hire security researchers, expand U.S. and European sales and strengthen its continuously verified ecosystem for agent tools.
In our report covering the AI safety market, we look at what the winners do differently, so you can learn from the best instead of guessing.
Arga Labs raised $10M in August 2026 to create realistic software environments where enterprise AI agents can be tested safely.
When was it?
Arga Labs announced the seed round on August 26, 2026.
Who are they?
Arga Labs builds stateful digital twins of enterprise applications so developers can train and test autonomous agents without exposing real production systems.
Geographical focus?
Arga Labs is based in San Francisco and builds infrastructure for enterprise AI developers across global software environments.
Why do we include them in the AI safety market?
Arga Labs lets companies evaluate autonomous behavior before deployment, placing it in AI agent simulation, evaluation and safety testing.
What is the company stage?
Arga Labs is at the seed and very-early-commercial stage, with a working platform for realistic enterprise agent testing.
How much did they raise?
Arga Labs raised $10 million.
What round is it?
Arga Labs raised a seed round.
Why did they raise?
Arga Labs raised the funding to expand its library of software simulations, improve their fidelity and grow the engineering organization.
Alice raised $140M in August 2026 to scale adversarial testing and guardrails for advanced AI models and agents.
When was it?
Alice announced the financing on August 25, 2026.
Who are they?
Alice, formerly ActiveFence, uses adversarial intelligence to stress-test AI models and agents and build defenses against jailbreaks, prompt injection and harmful behavior.
Geographical focus?
Alice was founded in Israel and serves major AI developers and enterprise customers globally.
Why do we include them in the AI safety market?
Alice directly tests and mitigates AI-specific vulnerabilities and harmful behavior, placing it in adversarial evaluation, red teaming and safety guardrails.
What is the company stage?
Alice is at the growth stage with strong product-market fit and was approaching $100 million in annual recurring revenue around the financing.
How much did they raise?
Alice raised $140 million.
What round is it?
Alice described the transaction as a growth funding round, while some market sources classify it as a Series C.
Why did they raise?
Alice raised the funding to advance its AI security platform, expand its proprietary adversarial-intelligence capabilities and scale commercial operations.
In our research about AI safety, we don't want to repeat what boring consulting reports say. Instead, we show you the real and latest market signals that matter.
Written by the New Market Pitch teamWe track new markets so founders and investors can move faster. About us
