Introduction
Redpine, a Stockholm-based AI data infrastructure startup, has secured €6.8 million in fresh funding led by NordicNinja, with participation from Luminar Ventures and node.vc, alongside several prominent technology founders and operators. The round brings the company's total capital raised to €9 million since its founding in 2024. The funding will be used to accelerate international expansion and grow its network of exclusive data partnerships.
The company was founded by Anders Hammarbäck, formerly of McKinsey and Antler, and David Österdahl, a veteran of Spotify and iZettle. Leonora Vesterbacka, who holds a PhD from CERN and previously worked in AI R&D at KBLab, serves as the founding data scientist. Together, they have built a platform that addresses one of the most pressing challenges in artificial intelligence: the sourcing of high-quality, legally compliant training data.
The Problem with Scraped Data
Most AI systems today are trained on data scraped from the internet. This approach is legally fragile, offers uncertain quality, and provides no competitive differentiation—everyone can access the same public datasets. Moreover, rights holders whose work is used receive no compensation, leading to growing legal pressure. Recent cases highlight the risks: Anthropic settled a $1.5 billion copyright lawsuit over book data, and new EU disclosure requirements are tightening the regulatory environment for AI training data.
The AI industry's data problem is structurally similar to the music industry's piracy crisis before Spotify. Piracy was widespread because licensed alternatives were cumbersome and expensive. Spotify succeeded by making licensed access easier and better than the illegal alternative. Redpine's founders explicitly frame their platform as the 'Spotify model for data'—not by making unauthorized scraping illegal, but by offering a licensed alternative that is superior in speed, quality, and cost.
The Spotify Model for Data
Redpine operates as a headless API layer that AI agents can query to retrieve premium licensed datasets in real time. Access is billed on a token-based usage model, meaning costs scale with actual consumption rather than a flat subscription fee. This pay-per-use approach allows developers to integrate high-quality data into their AI workflows without upfront commitments or complex licensing negotiations.
The platform evaluates data quality in real time, filtering out outdated or unreliable information before it reaches the agent. This is particularly critical for AI systems that perform multi-step reasoning, where inaccurate data can compound errors. The focus is on mission-critical domains: healthcare, legal, financial markets, scientific research, and news—areas where data accuracy can have life-or-death or multi-million-dollar consequences.
Redpine already provides access to more than 100 billion tokens of premium licensed data. The company positions itself as an API-native, agent-first alternative to traditional data annotation services. Competitors like Scale AI, Appen, and Defined.ai are built around human labeling workflows and static datasets. In contrast, Redpine delivers licensed data dynamically, in real time, through a simple API call—making it far more suitable for modern AI agents that need fresh, curated information on the fly.
Targeting High-Stakes Industries
The startup's initial focus on healthcare, legal, financial markets, scientific research, and news reflects a deliberate strategy. These sectors have strict regulatory requirements, high stakes for accuracy, and a willingness to pay for premium data. For example, an AI agent used in legal research must retrieve only verified court rulings and statutes; a healthcare diagnostic agent needs access to peer-reviewed studies and patient records under HIPAA compliance. Redpine's quality filtering and licensing guarantees address these needs directly.
The market for AI training data is growing rapidly, with Redpine citing a 24.9% annual growth rate within the broader trillion-dollar AI market. By offering a licensed alternative, the company aims to capture a share of this growth while helping enterprises mitigate legal risks associated with scraped data. The company's stated goal is to become the global category leader in AI data infrastructure within three to five years.
Strong Backing from Industry Insiders
The seed round attracted notable angel investors from leading AI companies, reflecting the industry's interest in solving the data sourcing problem. Investors include Colin M. Evans from OpenAI, Gustav Lindqvist from Perplexity, Anna Nordell Westling from Sana, and Daniel Langkilde, founder of Kognic (sold to Volvo). Several Spotify alumni also participated. New investor Peter Sarlin, co-founder and former CEO of Silo AI—the Finnish AI lab acquired by AMD for $665 million in 2024—adds further credibility.
This investor mix signals that both AI developers and enterprise customers see the value in a licensed data API. The participation of former Spotify employees is particularly fitting, given the company's 'Spotify for data' analogy. Just as Spotify transformed music consumption, Redpine aims to transform how AI agents access and pay for data.
The Road Ahead
Redpine's biggest challenge will be competing against well-capitalized US incumbents like Scale AI, which has raised over $1 billion. However, the legal environment may shift the balance. As regulators push for transparency and compensation in AI training, enterprises and AI labs may increasingly seek licensed data infrastructure. Rights holders, from publishers to research institutions, may also prefer to distribute through a platform like Redpine rather than negotiating direct deals with every AI developer.
Whether a Stockholm startup can win this market depends on execution speed, partnership depth, and the willingness of the ecosystem to adopt API-native solutions. With €9 million in total funding, a clear vision, and strong industry backing, Redpine is well-positioned to ride the wave of regulatory change. The next 12 to 18 months will reveal whether its platform can deliver on the promise of making licensed data as easy to access as streaming music.
Source: TNW | Startups-Technology News