AI Law
AI Structuring & Investments: IP Holding, SPV, and Corporate Architecture for AI Companies
An investor opens a Series A process. Due diligence starts. They find the IP is in the operating company, the founder’s pre-incorporation code was never assigned, and a contractor owns a core dataset. The round stalls. We design corporate architectures that prevent this — before the investor asks the question.
4–8 weeks
Typical engagement
EU · UAE · Switzerland · Global
Jurisdictions covered
Startups · VC Funds · Corporates
Who we work with
Why AI companies need a different structure
IP is your most valuable asset — and the most exposed
AI models, training datasets, and proprietary code are worth more than the operational business in most AI companies. Holding those assets in the operating company — where they're exposed to employment disputes, customer claims, and insolvency risk — is a structural mistake that's expensive to correct later.
Investors expect a clean structure
VC and PE investors have standard expectations: a holding company in a predictable jurisdiction, clear IP ownership, no conflicting claims from founders or contractors, and a structure that allows new investors to enter without reorganisation. Companies that arrive at Series A without this lose negotiating leverage.
Spin-offs and exits require preparation
Separating an AI division from a larger corporate, or preparing for a strategic sale, requires knowing exactly where the IP sits, which employees have assignment obligations, and how client contracts transfer. This work takes months — and costs more when done under time pressure.
The round is open and you have 8 weeks
Restructuring under active investor scrutiny is the most expensive way to do it. Investors won’t pause a deal process while you transfer IP, clean up contractor assignments, and reorganise entities. We work to defined timelines — and we have structured companies under live deal pressure before.
Common AI company structures: how they compare
A practical comparison of the most common corporate architectures used by AI companies at different stages.
The right structure depends on your stage, investor profile, jurisdiction mix, and exit scenario. We assess all four before recommending an architecture.
What's included
✓
Business model and value chain analysis (where IP is created and where it sits)
✓
Corporate structure design (holding, IP holding, operating entities)
✓
Jurisdiction selection for holding and IP holding companies
✓
IP transfer roadmap (assignment agreements, novation of contracts)
✓
Intra-group licensing structure (royalty model, sublicensing restrictions, change-of-control triggers)
✓
Employee and contractor IP assignment review and remediation
✓
Investor memo on structure, IP ownership, and risk allocation
✓
SPV design for VC/PE fund AI investments
✓
LP-facing structure documentation (diagrams, legal summaries)
✓
Co-investor and family office participation mechanics
✓
Spin-off and carve-out structuring for corporate AI divisions
✓
Strategic investor and M&A preparation (structure presentation, IP chain documentation)
ℹ️ We work with AI startups preparing for institutional rounds, VC funds structuring AI investments and SPVs, and corporates separating AI divisions into standalone entities.
How it works
Step 01
Current state mapping
Week 1
We map where your IP currently sits — which entity owns models, code, datasets, and brand — and identify conflicts: employment agreements without assignment clauses, contractor work without IP transfer, client contracts that may complicate restructuring.
Step 02
Structure design
Weeks 2–3
We design the target architecture: which entities to create, where to domicile them, how IP flows between them, and what the intra-group licensing model looks like. We model the structure against your investor profile and exit scenarios.
Step 03
Documentation and transfer
Weeks 3–6
We draft or coordinate: IP assignment agreements, intra-group license agreements, employee and contractor amendments, novation of key client contracts, and the investor-facing memo describing the structure.
Step 04
Investor readiness
Weeks 6–8
We prepare the materials investors will ask for: structure diagram, IP ownership chain, licensing terms, and answers to standard due diligence questions. For SPV transactions, we prepare fund documentation and LP-facing summaries.
How we've helped clients
AI Startup · Germany · Series A
IP holding structure for a Series A AI startup
Context
AI SaaS startup with distributed R&D team (Germany + Eastern Europe), EU and US clients. Key assets: proprietary models and datasets. Investors required clean IP ownership before closing the round.
→
Structure designed: holding + IP holding + German operating GmbH
→
IP transfer roadmap: assignment agreements, license from IP holding to operating entity
→
Licensing structure: royalty model, sublicensing restrictions, change-of-control triggers
→
Investor memo: structure, IP risk allocation, future round mechanics
⏱ 4–6 weeks
Outcome: investor-ready structure, clean IP chain
VC Fund · Luxembourg
SPV architecture for AI fund investments
Context
European VC fund making several large AI infrastructure investments, some via side-SPVs with co-investors and family offices. LP transparency and IP protection requirements across multiple deals.
→
Master fund + deal-specific SPV architecture designed
→
LP and co-investor rights clearly separated at SPV level
→
IP protections embedded in investment and shareholder agreements
→
LP-facing structure documentation: diagrams and legal summaries
⏱ 3–5 weeks
Outcome: standardised SPV model, reduced LP questions
Corporate · Switzerland
AI division spin-off and IP carve-out from a technology holding
Context
Swiss technology holding separating its AI platform into a standalone business for strategic investment. IP and team spread across multiple entities and jurisdictions. Needed clean carve-out before approaching investors.
→
IP ownership map across all current entities and jurisdictions
→
New structure: AI holding + regional operating entities with intra-group licenses
→
IP transfer plan: assignment agreements, novation of client contracts, employee amendments
→
Strategic investor memo: IP chain, cash flows, governance structure
⏱ 6–8 weeks
Outcome: clean AI business separation, investor-ready
Frequently asked questions
Why should AI IP be held in a separate entity from the operating company?
+
What jurisdiction should I use for an AI IP holding company?
+
What is an IP assignment agreement and do I need one?
+
How does an intra-group AI license work?
+
What is a side-SPV and when is it used in AI investments?
+
How do we handle IP that was created by contractors without assignment agreements?
+
What do investors look for in an AI company's corporate structure?
+
How long does it take to restructure a company before a fundraising round?
+
Related services
From the blog
Fundraising is coming. Is your structure investor-ready?
We assess your current IP position and identify structural gaps in one call — before your investor does it in DD. Restructuring from 4 weeks.
Or email us directly: info@wcr.legal