Paris-based Arlequin AI, which is developing proprietary models based on topological neural networks, raised a €28M Series A co-led by redalpine and OTB
Arlequin AI has secured Series A funding to develop its topological neural network architecture and expand the deployment of AI tools …
Context & Ripple Effects
French AI startups have previously raised for approaches positioned outside data-intensive mainstream models: AnotherBrain raised a €19M Series A in 2019 around lower-energy software that did not rely on large datasets. Arlequin brings that architectural thread back into the funding market with a larger early-stage round.
Paris AI financing in 2026 has also backed application-layer companies, including procurement and finance software maker Pivot and security-operations developer Qevlar AI. Arlequin is a different wager: proprietary model architecture rather than an AI workflow product.
First-order effects
- Arlequin gains €28M to develop its topological neural-network architecture and expand deployment of its AI tools, while co-leads redalpine and OTB take a direct stake in that technical approach.
Second-order effects
- The round gives architecture-focused developers such as AnotherBrain a clearer funding comparator, while investors can weigh proprietary-model bets against the better-defined workflow and security AI companies funded in Paris during 2026.
Third-order effects
- If Arlequin converts its architecture into deployed tools, European AI funding may broaden from application-layer companies toward companies seeking to own differentiated model architectures; deployment, rather than the novelty of the network design alone, will determine whether that shift holds.
The trend: European AI investors are backing a wider stack, pairing capital for workflow and agent products with selective bets on proprietary alternatives to prevailing model architectures.