Telekom and NVIDIA Launch Industrial AI Cloud: Munich’s 10,000‑GPU Cluster Redefines the Game for DACH CIOs
Eva Mickler
Deutsche Telekom and NVIDIA announced the launch of the Industrial AI Cloud at the end of April 2026. ...
Read Article
8 min Reading Time
German startups raised approximately €8.4 billion in venture capital in 2025 – 19 percent more than the previous year. At the same time, corporations are increasingly launching their own ventures rather than merely investing externally. According to McKinsey, 58 percent of experienced venture builders rank corporate venture building among their top-five strategic priorities. The reason? Traditional CVC (Corporate Venture Capital) delivers financial returns – but not strategic control. To shape tomorrow’s business fields, companies must build them themselves.
TL;DR
- €8.4 billion in venture capital flowed into Germany in 2025 – up 19 percent from 2024 and the third-highest total on record. 3,568 new startups launched – a record high (Startbase 2025).
- 58 percent prioritize venture building among their top five strategic goals: McKinsey reports that experienced venture builders are significantly expanding investments in building new business models (McKinsey 2025).
- 13× higher prioritization: Companies with venture-building experience are 13 times more likely than others to further scale their venture activities.
- Siemens Next47, Bosch Ventures, TRUMPF Ventures as German role models: Climate Tech, Industry 4.0, and AI as focal areas for corporate-founded ventures.
- Build – not just invest: CVC delivers financial returns; corporate venture building delivers strategic control over emerging business fields.
Why CVC Alone Is No Longer Enough
Corporate Venture Capital has long been the go-to instrument for corporations seeking innovation beyond their core operations. Siemens, Bosch, SAP, and Deutsche Telekom have run CVC arms for years, investing in external startups. The problem? A minority stake in a startup grants insight – but not control.
A growing strategic insight among executives: To claim a new business field, you must build it yourself – not as a slow-moving internal innovation unit (hindered by corporate logic), but as an independent venture with its own governance, team, and mandate.
McKinsey calls this Serial Venture Building: systematically and repeatedly launching new ventures, rather than making one-off investments in external startups. Data shows companies with venture-building experience are 13 times more likely to further elevate venture building as a priority. It’s a flywheel effect: each successful venture makes the next one more likely.
Source: Startbase / German Startup Association, 2025
Three Models: How Corporations Build Ventures
Model 1: The Internal Venture Builder. A dedicated team inside the corporation identifies business opportunities, validates hypotheses, and launches ventures. Example: Siemens Next47, which has systematically built deep-tech ventures since 2016. Advantage: Access to corporate resources (customers, data, infrastructure). Disadvantage: Risk of corporate inertia slowing execution speed.
Model 2: The External Venture Builder (Studio). The corporation engages an external venture studio to design and launch ventures. Over 20 specialized studios operate in Germany today. Advantage: Startup speed and culture. Disadvantage: Weaker strategic integration into the core business.
Model 3: The Joint Venture with Startup DNA. The corporation co-founds a new company alongside an experienced founding team. The corporation contributes market access, customers, and infrastructure; the founding team brings operational velocity and product development expertise. Advantage: Best-of-both-worlds synergy. Disadvantage: Complex governance balancing corporate and startup logic.
“Companies that launched new ventures in the past five years are 13 times more likely than others to further elevate venture building as a strategic priority. Serial building creates a flywheel effect.”
McKinsey, The Way to Win in Corporate Venturing: Serial Building and AI, 2025
AI as Accelerator: Why 2026 Is the Right Moment
McKinsey’s analysis carries the subtitle “Serial Building and AI” – for good reason. AI is fundamentally transforming venture building. Hypotheses can be validated faster. Prototypes emerge in days – not months. And scaling AI-native business models demands far fewer people than traditional ventures.
For executives, this means: The cost and risk of venture building are falling. An AI-native venture can reach market readiness with a team of five to ten people – where five years ago, 30 to 50 were required. That changes the ROI calculus entirely.
At the same time, AI is spawning entirely new business fields ripe for corporate capture: AI-powered quality control in manufacturing, autonomous logistics optimization, and predictive maintenance-as-a-service. These domains demand deep domain expertise – something established corporations possess, but pure-play tech startups often lack.
Success Factors: What Winners Do Differently
1. Independent Governance. Successful corporate ventures operate with their own board, dedicated budget, and autonomous decision-making pathways. Every loop back to corporate committees slows progress. The board sets strategic direction; the venture team owns operational execution.
2. Dedicated Founder. A corporate venture needs a founder – not a corporate manager juggling the initiative alongside daily duties. It requires someone fully committed, with entrepreneurial DNA. The most common cause of corporate venture failure? No one waking up each morning thinking solely about that one venture.
3. Define Kill Criteria. Not every venture will succeed. Top-tier venture builders set clear criteria upfront for when to shut down: e.g., no demonstrable product-market fit after 12 months triggers termination. Without explicit kill criteria, “zombie ventures” linger – tying up resources without delivering value.
4. Leverage Unfair Advantages. The corporation must give the venture something a standalone startup cannot replicate: customer access, data, industry knowledge, regulatory expertise, or infrastructure. If the venture gains no unfair advantage from its corporate parent, it could – and should – have launched independently.
5. Exit Strategy from Day One. Will the venture integrate into the core business? Spin out as an independent company? Be sold? The exit strategy shapes every decision – from legal structure to team composition. Without a clear vision for exit, the venture lacks direction.
Conclusion
Corporate venture building is not innovation theater. It’s the answer to how corporations claim new business fields – without relying on acquiring external startups. The data is unambiguous: €8.4 billion flowed into German startups in 2025. Corporations that build – not just invest – secure strategic control over tomorrow’s markets. The timing is ideal: AI lowers venture-building costs, the generational transition in Germany’s Mittelstand opens acquisition windows, and experienced venture builders accelerate momentum through the flywheel effect. The question for the board: Build – or watch?
Frequently Asked Questions
What distinguishes corporate venture building from CVC?
CVC (Corporate Venture Capital) invests in external startups and acquires minority stakes. Corporate venture building launches wholly owned ventures from scratch. CVC delivers financial returns and strategic insights. Venture building delivers strategic control over emerging business fields. The two approaches complement – but do not replace – each other.
How much does a corporate venture cost?
Typically €1-5 million for the validation phase (6-12 months) and €5-20 million to reach scalability (12-24 months). AI-native ventures may launch with smaller budgets. Crucially, the corporation must commit to shutting down the venture after 12 months if no product-market fit emerges.
Which industries suit corporate venture building?
Industries demanding deep domain expertise: manufacturing (Industry 4.0, predictive maintenance), financial services (RegTech, embedded finance), healthcare (digital health), energy (CleanTech), and logistics (autonomous optimization). The key: The corporation’s domain knowledge must constitute an unfair advantage – one that independent startups cannot replicate.
How long until a corporate venture becomes profitable?
Typically 3-5 years. Months 1-12 focus on hypothesis validation (product-market fit). Months 13-36 cover build-out and initial scaling. By Year 3, the venture should deliver measurable business value – whether via revenue, strategic positioning, or data assets.
What happens to failed ventures?
Professional venture builders expect a 30-50 percent success rate. Even failed ventures deliver value: market insights, reusable technology components (integrated into the core business), and teams enriched with founder experience. The biggest mistake? Treating failures as defeats – not as investments in organizational learning.
Further Reading
Platform Ecosystems: Build, Buy, or Join
Boardroom Generational Transition: 545,000 Successors Sought
AI Is Not a Tech Problem: 27 Percent of CEOs Fail
More from the MBF Media Network
MyBusinessFuture: Reboot Germany – €735 Billion, Three Mittelstand Champions
cloudmagazin: Platform Engineering 2026
SecurityToday: Agentic AI Security – When AI Agents Become Attack Targets
Header Image Source: Pexels / RDNE Stock project (px:7413910)