Humanos
Risk and authorization infrastructure for autonomous AI agents, providing continuous risk scoring and passports.
Website: https://www.humanos.tech/
Open sources
| Attribute | Detail |
|---|---|
| Name | Humanos |
| Tagline | Risk and authorization infrastructure for autonomous AI agents, providing continuous risk scoring and passports. |
| Headquarters | Lisbon, Portugal |
| Founded | 2025 |
| Stage | Seed |
| Business Model | API / Developer Platform |
| Industry | Security |
| Technology | AI / Machine Learning |
| Geography | Western Europe |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding Label | Seed (total disclosed ~$3.2M) |
Links
Open sources
- Website: https://www.humanos.tech/
- LinkedIn: https://www.linkedin.com/company/humanos-tech
What an Investor Needs First
Open sources
Humanos is building the risk and authorization infrastructure for autonomous AI agents, a bet that the operational safety of these systems will become a critical, regulated market as agent adoption accelerates [SesaMers, September 2026]. The Lisbon-based startup, founded in 2025, provides a live 0-100 risk score and a 'risk passport' for each agent, creating a shared network where insurers and lenders can underwrite agentic activity [The SaaS News, September 2026]. Its initial wedge is serving regulated fintechs, insurers, and hospitals, where it reported over 350 deployments within a year of launch [Portugal Startup News, June 2026].
The founding team is described as seven repeat founders with a collective history of 19 startups and five exits, a depth of experience that is unusual for a seed-stage company in this category [Portugal Startup News, June 2026]. The company secured a $3.2 million seed round in September 2026, led by Anthemis with participation from Springbank and founders of Sword Health and Feedzai, signaling institutional validation for its infrastructure approach [SesaMers, September 2026]. Over the next 12-18 months, the key test will be whether it can convert early technical deployments into recurring, high-value contracts and establish its risk passport as a standard for agent-to-agent trust.
Partially corroborated -- Core funding and deployment claims are reported by multiple outlets, but key product details and team background lack independent primary source verification.
Taxonomy Snapshot
| Axis | Snapshot |
|---|---|
| Stage | Seed |
| Business Model | API / Developer Platform |
| Industry / Vertical | Security |
| Technology Type | AI / Machine Learning |
| Geography | Western Europe (Lisbon, Portugal) |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding | Seed (total disclosed ~$3,200,000) |
Inside the Company
Open sources
Humanos Labs Inc. was founded in 2025 and operates from Lisbon, Portugal, as an AI risk infrastructure provider [Humanos]. The company's public emergence was marked by a launch announcement in June 2026, which described a product already deployed across more than 350 regulated organizations [Portugal Startup News, June 2026]. This was followed by a $3.2 million seed financing round in September 2026, led by venture firm Anthemis [SesaMers, September 2026].
The founding team is unusually large, comprising seven individuals identified as co-founders. Public reports characterize this group as repeat founders, though detailed independent verification of prior ventures and exits for most members is not yet available in widely cited sources [Portugal Startup News, June 2026]. CEO Pedro Andrade is a named Forbes 30 Under 30 honoree from 2021, and his LinkedIn profile confirms employment with Humanos beginning in March 2025 [Forbes, July 2021] [LinkedIn].
Partially corroborated -- Key milestones (founding year, launch, funding) are confirmed by multiple publications, but team composition and prior venture claims rely on a single regional news source and founder-provided profiles.
Under the Hood
Reported and inferred
Humanos positions its core technology as a runtime authorization layer for autonomous AI agents, a system designed to intercept and evaluate agent actions before they are executed. The product, as described in public materials, assigns each agent a continuously updated risk score from 0 to 100 based on a composite of its identity, permissions, and historical behavior [PERPLEXITY SONAR PRO BRIEF]. This score is recorded in a persistent "risk passport," creating a portable audit trail. The operational interface is a Risk API that developers integrate to perform checks for identity, authorization, and guardrail compliance prior to an agent taking any action [PERPLEXITY SONAR PRO BRIEF]. The company's website frames this as turning static approvals and policies into "portable runtime authorization" that can be verified before execution and proven afterward [Humanos].
Public reporting indicates the technology is already integrated into specific customer workflows. InsureNow is cited as integrating Humanos signals into its underwriting processes, while Lusíadas, a Portuguese hospital group, Numo, and DataWhisper have reportedly rebuilt systems on the platform [Humanos-App/humanos-website-launch]. The integration with DataWhisper's SmartInsights.CortexOS is positioned as a combined governance and orchestration layer for regulated organizations [Lucy Underwood - Self-employed | LinkedIn]. These deployments suggest the product's initial wedge is providing a behavioral risk signal for third-party financial and insurance products, effectively allowing lenders and insurers to underwrite agent activity.
The architecture appears to be a cloud-based API service, with pricing referenced as pay-for-execution rather than per-seat [Humanos]. While the company's own materials and launch coverage describe the scoring and passport system, the specific algorithms, data sources for the risk model, and details on real-time enforcement mechanisms are not publicly disclosed. The claim of "more than 350 deployments" is repeated across multiple news outlets but is not independently verified with a customer list [SesaMers, September 2026] [eu.vc].
Open sources
A market for autonomous AI agent security is emerging from a specific, high-stakes customer need: regulated organizations cannot deploy these systems at scale without a verifiable, auditable record of operational risk. The demand is not for general AI governance but for runtime authorization and underwriting signals that translate directly into compliance and capital decisions.
Quantifying the total addressable market for AI agent risk infrastructure is premature, as the category itself is nascent. However, the initial customer segments are well-defined and carry significant spending power. Humanos targets fintechs, insurers, and hospitals, industries where regulatory scrutiny and liability exposure are acute. A useful analog is the broader AI governance, risk, and compliance (GRC) software market, which Gartner estimated at $2.5 billion in 2025 and projected to reach $5.5 billion by 2028 (analogous market, Gartner). While this includes a wider set of tools, it indicates the budget and urgency large enterprises attach to AI risk management.
The primary demand driver is the accelerating, yet cautious, adoption of autonomous agents for high-value workflows in regulated sectors. Agents handling loan approvals, insurance claims, or patient data interactions create new, dynamic risk profiles that static compliance frameworks cannot address. This creates a wedge for a product that can produce a continuous, evidence-based risk score. A secondary driver is the potential for these risk signals to feed into adjacent financial markets, such as underwriting insurance for AI operations or extending credit to agent-run businesses, as hinted at in Humanos's reported marketplace ambitions [Portugal Startup News, September 2026].
Key adjacent markets include traditional application security, identity and access management (IAM), and insurance technology. Humanos's proposition differs by focusing on the behavioral patterns of non-human actors in real-time. The regulatory environment is a significant tailwind; legislation like the EU AI Act imposes strict requirements for high-risk AI systems, mandating risk assessments and human oversight. This regulatory pressure effectively mandates the kind of continuous monitoring and documentation that Humanos aims to provide.
AI GRC Software Market 2025 | 2.5 | $B
AI GRC Software Market 2028 | 5.5 | $B
The projected near-doubling of the broader AI GRC market by 2028 suggests strong underlying demand for risk management tooling, though it does not isolate the agent-specific segment Humanos occupies.
Partially corroborated -- Market sizing is based on an analogous sector report from a major analyst firm. Demand drivers and regulatory forces are well-established industry trends, but the specific market for agent risk scoring is not yet independently sized.
Competition and Substitutes
Reported and inferred Humanos enters a nascent, fragmented market for AI agent governance, where its primary competition is not a single incumbent but a collection of point solutions and architectural approaches that address different slices of the trust problem.
The structured research did not surface any directly named competitors for the company's specific offering of continuous risk scoring and authorization passports for autonomous agents. This absence is itself a data point, suggesting the category is either poorly defined or that Humanos is among its earliest entrants. Without a named competitive set, a detailed comparison table cannot be constructed. Competitive analysis must therefore be based on adjacent categories and inferred market structure.
From a segment-by-segment view, the competitive map is defined by what is being secured. For securing the agentic workflow itself, competitors include agentic orchestration platforms like DataWhisper's SmartInsights.CortexOS, with which Humanos integrates [Humanos]. These platforms provide governance and orchestration but may lack the specialized, cross-platform risk scoring that Humanos offers as a dedicated service. For securing the underlying models, the field includes traditional AI safety and alignment research, which operates at a more fundamental, pre-deployment level. For securing the infrastructure layer, competition could come from cloud providers' native security tooling or API security companies that might extend their scope to agent behavior. The most direct adjacent substitutes are likely internal, custom-built authorization systems within large regulated enterprises, which Humanos aims to replace with a standardized network.
Humanos's defensible edge today appears to be its early focus on turning agent behavior into a financial signal. The reported integrations with insurers like InsureNow for underwriting and lenders for credit assessment position its 'risk passport' as more than a security log; it is a reputational and financial identity [Humanos-App/humanos-website-launch, SesaMers, September 2026]. This edge is durable if the company can establish its scoring methodology as a de facto standard among financial institutions, creating network effects. However, it is perishable if a larger player in insurance tech or payment processing decides to build a similar scoring layer and leverages its existing customer relationships to adopt it faster.
The company's most significant exposure lies in its reliance on a nascent market. If the adoption of autonomous AI agents in high-stakes, regulated workflows proceeds more slowly than anticipated, the entire category could remain a niche. Furthermore, Humanos is exposed to competition from below by open-source frameworks that offer basic permissioning, and from above by large cloud security suites (e.g., from Microsoft or Google) that could eventually bundle agent risk monitoring as a feature. The company does not currently own the distribution channel into enterprise security teams, a gap that could be exploited by established cybersecurity vendors.
The most plausible 18-month scenario involves continued category definition and early consolidation. A winner will emerge if a clear use case,such as AI agent underwriting for commercial insurance,achieves rapid, tangible ROI, drawing in capital and partnership deals. A loser in this segment would be any company that fails to move beyond scoring to providing enforceable, real-time authorization that directly prevents costly agent errors or fraud. For Humanos, the path to winning hinges on converting its reported 350+ deployments into deep, embedded workflows that generate continuous scoring data, thereby making its network more valuable than any single-point alternative.
Partially corroborated -- Competitive analysis is inferred from product positioning and adjacent markets; no direct competitors were named in sourced material.
Opportunity
Open sources
The core bet for Humanos is that the economic activity of autonomous AI agents will become a multi-trillion-dollar flow requiring its own dedicated, real-time risk infrastructure, a market that does not meaningfully exist today.
The headline opportunity is for Humanos to become the de facto credit and insurance bureau for the agentic economy, analogous to what Experian or S&P Global became for human-driven commerce. The company's early wedge, turning agent behavior into an underwriting signal for insurers and lenders, positions it at a critical transactional chokepoint [PERPLEXITY SONAR PRO BRIEF]. If AI agents are to autonomously transact, secure credit, or be insured, a trusted, third-party risk assessment layer becomes mandatory infrastructure. The company's reported traction with over 350 regulated organizations, including named deployments with insurers like InsureNow and hospitals like Lusíadas, provides initial evidence that this need is being recognized by first-mover customers in high-stakes verticals [SesaMers, September 2026]. This is not a feature for a developer platform; it is a foundational utility for a new asset class.
Growth could follow several distinct, high-conviction paths, each with identifiable catalysts.
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Regulatory Standard-Bearer | Humanos' risk passport becomes a mandated compliance artifact for AI agents in regulated industries (finance, healthcare). | A major financial regulator (e.g., ECB, SEC) or industry body issues guidance requiring demonstrable, auditable risk controls for autonomous systems. | The company is already engaging with regulated entities that face strict compliance burdens, and its product is framed as providing provable audit trails [Humanos]. |
| Embedded Risk Layer | The Risk API becomes a white-label component embedded within every major agentic platform (e.g., DataWhisper's CortexOS) and cloud provider's AI suite. | A strategic partnership with a leading AI orchestration provider or cloud hyperscaler is announced. | An integration with DataWhisper's SmartInsights.CortexOS is already cited, demonstrating the model of becoming a specialized component within a broader stack [Humanos]. |
| Marketplace Dominance | The planned marketplace for agent insurance, credit, and yield products becomes the primary liquidity layer, with Humanos taking a transaction fee on all facilitated economic activity. | The company onboards a critical mass of capital providers (insurers, lenders) and agent developers, creating a two-sided network. | The September 2026 financing report explicitly describes building a marketplace as a next step, with backing from fintech-focused investors like Anthemis [Portugal Startup News, September 2026]. |
Compounding for Humanos is driven by a classic network-and-data flywheel. Each new agent deployment generates more behavioral data, improving the accuracy and predictive power of the risk model. A more accurate model attracts more insurers and lenders to the platform, offering better terms. Better terms, in turn, attract more agent developers seeking capital and coverage, which further expands the data network. Early signs of this flywheel are suggested by the company's claim of targeting "$1 billion in agentic economic activity" flowing through its systems, implying a focus on transaction volume that feeds the model [Portugal Startup News, September 2026]. The risk passport itself could become a portable reputation score that locks in users, as switching would mean abandoning a valuable historical record.
The size of the win, should the company capture this nascent category, is anchored in the valuation of established financial data and analytics platforms. For a scenario where Humanos becomes the central risk infrastructure for a significant portion of agentic activity, a reasonable comparable is S&P Global's Market Intelligence division, which trades at approximately 10x forward revenue for essential risk and compliance data. If Humanos were to facilitate even a single-digit percentage of a future trillion-dollar agentic economy and capture basis points in fees, the revenue potential would support a multi-billion dollar valuation (scenario, not a forecast). The $3.2 million seed round provides the initial capital to attempt the first turn of the flywheel [SesaMers, September 2026].
Partially corroborated -- Core traction claim (350+ deployments) is reported by multiple outlets but not independently verified. The product mechanics and growth scenarios are extrapolated from company and investor statements.
Sources
Open sources
[SesaMers, September 2026] Humanos raises $3.2M seed for AI agent risk scores | https://www.sesamers.com/funding/humanos-raises-3-2m-seed-ai-agent-risk-score/
[The SaaS News, September 2026] Humanos Raises $3.2M Seed | https://www.thesaasnews.com/news/humanos-raises-3-2m-seed/
[Portugal Startup News, June 2026] Humanos launches to enforce human authorization before AI acts | https://portugalstartupnews.com/2026/06/03/humanos-launches-to-enforce-human-authorization-before-ai-acts/
[Portugal Startup News, September 2026] Humanos raises $3.2M to build risk infrastructure for AI agents | https://portugalstartupnews.com/2026/09/28/humanos-raises-3-2m-to-build-risk-infrastructure-for-ai-agents/
[Humanos] The Risk Network for AI | https://www.humanos.tech/
[Forbes, July 2021] “Unknown Amazon” Investigates How Amazonia’s Indigenous People, Communities Feel About Climate Change and Progress | https://www.forbes.com/sites/joesills/2021/07/15/unknown-amazon-investigates-how-amazonias-indigenous-people-communities-feel-about-climate-change-and-progress/
[LinkedIn] Pedro Andrade - Talkdesk | https://www.linkedin.com/in/pedromandrade/
[Humanos-App/humanos-website-launch] Case study page (via raw research) | https://www.humanos.tech/case-studies/ralio
[Lucy Underwood - Self-employed | LinkedIn] Post referencing DataWhisper's SmartInsights.CortexOS | https://www.linkedin.com/in/lucy-underwood-ai/
[eu.vc] Article referencing deployments (via raw research) | https://eu.vc/
[PERPLEXITY SONAR PRO BRIEF] Web-grounded research brief compiled from public sources | (Research synthesis; primary sources cited individually above)
[Gartner] AI governance, risk, and compliance (GRC) software market estimates | (Analyst report referenced for analogous market sizing)
Articles about Humanos
- Humanos Scores 350 Deployments With a Risk Passport for AI Agents — The Lisbon startup is selling a 0-100 risk API to regulated fintechs and hospitals, backed by a $3.2 million Anthemis-led seed round.