Observed Signal · Jul 13, 2026 · Analysis · Source: t3n · Impact: 3/5 · Sentiment: Negative
Only 1 in 10 German Firms Use EU AI Models
An analysis by European AI platform GreenPT of 514 identifiable German AI companies finds strong dependence on US AI providers: among firms that publicly name a model provider, about 53% rely on US companies such as OpenAI, Anthropic, Google, Microsoft or AWS. When restricted to companies that list a provider on their website (185 firms), the US-dependency rises to roughly 60%. Mistral — the only EU-based frontier lab mentioned — is used by about 10% of companies that disclose a model. The report also highlights a sovereignty gap: 173 firms market EU/Germany data residency or on‑premise deployment, yet 72 of those still use proprietary US models, especially in compliance-sensitive sectors.
Reveals a significant European sovereignty gap in AI model usage and widespread reliance on US providers, which affects data residency, compliance risk, and vendor choice for companies in AdTech/MarTech and other sectors.
Track OpenAI Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- GreenPT analyzed 514 publicly identifiable German AI companies; 221 disclose which AI model they use.
- About 53% of German AI companies that publicly name a provider rely on US companies (OpenAI, Anthropic, Google, Microsoft, or AWS).
- Among 185 companies that name a provider directly on their website, US-dependency rises to ~60%.
- Mistral is used by roughly 10% of companies that disclose their model; OpenAI models are the most used at about 41% among disclosing firms.
- 173 of the 514 companies advertise EU/Germany data residency or on-premise deployment, but 72 of those still use proprietary US models.
Connected Companies & Entities
10 Entities mapped“Beim Blick auf die verwendeten KI-Modelle zeigt sich allerdings, dass der Großteil nach wie vor auf Anbieter wie OpenAI oder Anthropic setzt...”
“Beim Blick auf die verwendeten KI-Modelle zeigt sich allerdings, dass der Großteil nach wie vor auf Anbieter wie OpenAI oder Anthropic setzt...”
“Von den deutschen KI-Unternehmen, die öffentlich einen Anbieter nennen, sind rund 53 Prozent von US-Unternehmen wie OpenAI, Anthropic, Googl...”
“Von den deutschen KI-Unternehmen, die öffentlich einen Anbieter nennen, sind rund 53 Prozent von US-Unternehmen wie OpenAI, Anthropic, Googl...”
“Mistral, das einzige in der EU ansässige Frontier-Labor, erreicht nur etwa jedes zehnte Unternehmen, das Angaben zu seinem KI-Modell macht....”
“Nur Microsoft Copilot und Meta Llama sind noch seltener vertreten....”
“Hier findest du externe Inhalte von TargetVideo GmbH, die unser redaktionelles Angebot auf t3n.de ergänzen....”
“Europäische KI-Modelle bleiben hinter der US-Konkurrenz zurück. (Bild: Shutterstock/Samuel Boivin)...”
“t3n – digital pioneers | Home...”
“Von den deutschen KI-Unternehmen, die öffentlich einen Anbieter nennen, sind rund 53 Prozent von US-Unternehmen wie OpenAI, Anthropic, Googl...”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
Chinese AI Models Underused by German Companies Despite Performance
A new Bitkom survey reveals that German companies using AI overwhelmingly prefer US-based models, with 76% relying on OpenAI, while only 2% use DeepSeek and 1% use Alibaba's Qwen. Despite Chinese AI models performing competitively and being cheaper, concerns over data privacy and cybersecurity hinder adoption. The AI Index Report 2026 from Stanford shows the quality gap between US and Chinese models has narrowed, exemplified by the release of Moonshot AI's Kimi K3, which is claimed to rival leading US models. German industry leaders call for more support of European AI providers to foster digital sovereignty.
ifo: German Firms Highly Dependent on US Tech
An ifo Institute survey finds that nearly 88% of German companies use digital products and services from US providers and 31% consider themselves strongly dependent. Despite broad risk awareness, many firms take little action: over one in five see their dependency as risky but plan no countermeasures, and among those recognizing risk more than 40% do not intend to change. The survey was prompted by a temporary restriction on access to Anthropic's AI models ordered by the US Department of Commerce in June. Companies prefer switching to European suppliers or diversifying providers, but limited European compute capacity and infrastructure constrain practical moves. ifo calls for faster expansion of European data center and energy infrastructure and shorter permitting processes.
Low AI Budgets Slow Germany’s AI Transformation
A Strand Partners study for AWS, “Erschließung des KI‑Potenzials in Deutschland 2026,” finds 63% of German companies use AI but only 15% apply it widely enough to create new products, business models or workflows — down from 21% last year and below the European average (22%). Most firms remain at basic applications (chatbots, copilots) rather than next‑generation agentic or physical AI; only 22% initially know about agentic AI though 57% could imagine testing it after explanation. Firms using agentic AI report markedly stronger outcomes (92% see productivity gains; 96% expect growth; 48% report higher revenues). Key barriers are missing AI/digital skills (49%), lack of dedicated AI budgets (39%) and rising compliance costs (~44% of tech spend). The study warns Germany risks losing growth, investment and startups unless capital, talent and regulatory frameworks improve.
Track Real-Time Market Signals & Shifts
Set up custom watchlists to receive automated, evidence-grounded executive digests whenever material signals or shifts occur across your tracked landscape.