Maisa, a Spanish startup focused on building hallucination-resistant AI agents, announced today a $25 million seed investment led by Creandum, with participation from Forgepoint Capital through its European joint venture with Banco Santander. Both NFX and Village Global—early backers in Maisa’s $5 million pre-seed round in December 2024—also joined this round. The latest infusion positions Maisa as one of the most ambitious European challengers in enterprise AI.
A Breakthrough Year for Maisa
Maisa emerged in 2024 under the leadership of David Villalón (CEO) and Manuel Romero (CSO). Villalón had previously shaped AI product strategy at Clibrain and Voicemod, while Romero distinguished himself as one of the world’s most active open-source contributors on HuggingFace, releasing over 700 models with 15 million monthly downloads. Within a year of founding, Maisa landed on Gartner’s Hype Cycle for AI and Future of Work, listed alongside global giants like Google and Amazon. For the first time, a Spanish startup earned that recognition—an achievement signaling Maisa’s credibility in a crowded AI market.
The $25 million round fuels an aggressive growth agenda. Maisa plans to expand hiring across AI research, engineering, sales, and customer success, while scaling its presence in Europe and North America. The company aims to meet rising enterprise demand for AI systems that execute complex tasks with precision and full auditability.
Introducing Maisa Studio: AI Digital Workers for Everyone
Alongside the funding, Maisa unveiled Maisa Studio, a next-generation agentic process automation platform. Studio enables citizen developers—professionals with deep domain expertise but little IT background—to deploy powerful AI “digital workers” through natural language.
The idea is simple yet disruptive: onboarding a digital worker should feel as easy as onboarding a new colleague. A compliance officer, for instance, can describe a process in plain words—such as reconciling cross-border transactions or flagging supply chain risks—and Studio translates that into a repeatable, transparent workflow. These digital workers:
- Require no pre-built dataset or coding expertise.
- Learn through HALP (Human-Augmented LLM Processing), Maisa’s method of training agents on the job.
- Document every decision step in a Chain-of-Work, an auditable trail that shows how conclusions were reached.
This approach directly tackles one of AI’s most notorious problems: hallucination. Instead of guessing probabilistically, Maisa’s Knowledge Processing Unit (KPU) drives workers to follow logical, step-by-step reasoning.
Enterprise Adoption: From Pilots to Production
Unlike many AI platforms still stuck in proof-of-concept purgatory, Maisa’s technology already runs in production at major organizations.
- A global investment bank replaced its manual media screening process with Maisa workers. These agents extract facts from thousands of news articles, assess reputational risks, and generate audit-ready summaries in minutes.
- A large financial services firm automated transaction checking and reconciliation. Studio eliminated 99% of false positives and delivered a 10x productivity boost—all without engineering intervention, only three onboarding sessions.
Maisa positions these deployments not as experiments but as live operations inside some of the world’s most compliance-conscious industries. That credibility strengthens investor confidence that Maisa can scale beyond pilots into mission-critical workflows.
Why Enterprises Need Maisa’s Approach
Industry research underscores the challenge Maisa addresses. According to MIT’s NANDA initiative, 95% of generative AI pilots fail to deliver measurable financial impact. IDC reports 88% of pilots never reach production, while BCG estimates only 22% of companies move AI past proof-of-concept. At the same time, HR analysts at i4cp find 70% of firms struggle to equip staff with AI skills, and Bain notes 44% of executives cite lack of in-house expertise as a top barrier.
Maisa neutralizes these bottlenecks. By removing the need for developers, pre-labeled datasets, or extensive retraining, Maisa’s digital workers empower non-technical staff to implement automation themselves. The Chain-of-Work offers full transparency, reassuring compliance teams. And HALP ensures accuracy improves continuously through human-in-the-loop reinforcement.
Investor Confidence and Strategic Validation
David Villalón, Maisa’s co-founder and CEO, emphasized the trust factor in enterprise AI:
“This investment validates that AI in business must be built on trust. Our platform gives teams the power to use AI workers which not only perform with incredible accuracy and intelligence, but also explain and prove their logic, and document every step in the work process. It means users can scale AI at pace, do so safely and without the need for an entire development team to support.”
Peter Specht, General Partner at Creandum, highlighted the technical breakthrough:
“Maisa is solving one of the toughest challenges in AI: making it reliable and safe to use in mission-critical business operations. With a string of top global companies already using the technology and massive demand, we are extremely confident of seeing significant growth at scale.”
Alberto Yepez, representing Forgepoint Capital International, pointed to sector-specific relevance:
“Maisa’s platform is purpose-built for compliance-conscious industries like finance, where decisions must be traceable and outcomes consistent. The company’s early success demonstrates there’s real market pull for AI done the right way, and Studio will make it easy for anyone in an organisation to use digital workers quickly and effectively.”
Model-Agnostic, Enterprise-Ready Design
Maisa built its system to remain model-agnostic. That means the platform can integrate any general-purpose LLM—whether from OpenAI, Anthropic, or open-source communities—and transform it into a safe, specialized agent tuned for regulated industries. Customers can deploy Studio either in Maisa’s secure cloud or within private infrastructures to meet compliance requirements.
The Knowledge Processing Unit underpins this architecture. It elevates LLMs from “probabilistic guessers” into deterministic task executors. Every output includes a Chain-of-Work record, ensuring managers can inspect exactly how each step unfolded. This transparency makes AI adoption less risky in sectors like banking, automotive, and energy, where errors can have regulatory or financial consequences.
A Wider Vision: Equipping the Workforce
Maisa’s ambitions stretch beyond automation. The company aims to solve the AI skills gap by empowering non-technical professionals to deploy advanced automation themselves. A supply chain analyst, a financial auditor, or a compliance officer can create digital workers without waiting months for IT backlogs.
Studio integrates seamlessly with over 450 third-party systems out of the box. It can connect to any documented API, adapt to custom undocumented APIs, and trigger automations via web, email, API, or webhook. Updates and refinements happen inside the same conversational interface, meaning users continuously evolve workflows without technical bottlenecks.
In practice, this flexibility transforms AI from an experimental tool into a practical everyday asset. Companies no longer ask, “How can we build AI?” but instead, “Which processes should we automate next?”
Looking Ahead
Maisa’s trajectory reflects a broader shift in enterprise AI. The industry faces skepticism after waves of over-hyped pilots failed to produce returns. Maisa distinguishes itself by focusing on reliability, traceability, and real deployments. That pragmatic approach resonates with investors, as demonstrated by the high-profile participation from Creandum, Forgepoint, NFX, and Village Global.
With $25 million in fresh capital, Maisa now has resources to deepen its research bench, accelerate product adoption, and expand globally. Studio gives enterprises a clear pathway to scale AI without waiting for scarce technical expertise. If the early deployments serve as a preview, Maisa could emerge as the default platform for enterprises seeking trustworthy AI agents.
Also Read – 20 Side Hustles That Could Make You $10K a Month