The essential facts
The German government aims for at least 55% of SMEs to use AI by 2030. Its cross-ministry AI task force proposed 18 key measures. These are policy objectives—not achievements to date, guaranteed grant awards or a general mandate for every business to procure AI software.
What was announced at the 2026 Digital Summit?
The federal government published its AI Nation 2030 vision on 2 November. A cross-ministry task force outlined 18 key measures covering skills, safety, infrastructure and real-world adoption.
It set AI adoption targets of at least 55% for SMEs and 80% for large businesses by 2030. The report describes a strategy whose targets and measures will be reviewed as technology changes.
How widely is AI used in German businesses today?
Germany’s Federal Statistical Office reported that 26% of covered businesses used at least one AI technology in 2025. The share was 23% among businesses with 10–49 employees and 36% among those with 50–249 employees.
The survey covers businesses with at least ten employees. Its size classes and definitions are not necessarily identical to the government’s 55% goal. These figures must not be treated as a directly comparable measure of progress without checking the underlying populations.
Does the strategy require every company to adopt AI?
No. A national target does not impose a blanket requirement for each business to purchase an AI system. Specific GDPR and EU AI Act obligations may apply depending on the organisation’s role, data and use case.
Reviewing data handling, permissions and employees’ AI literacy is more useful than adopting a tool solely because AI is a policy priority.
Where should an SME begin?
Choose one repeated task with clear inputs and outputs: preparing customer responses, reviewing documents or searching approved internal knowledge.
Evaluate whether a conventional workflow, better interface or API could solve the problem more reliably. Not every digital challenge calls for a language model.
What belongs in the engineering risk assessment?
A practical design accounts for data flows, access rights, vendors, logging, fallback procedures and human approval. Confidential data must not be shared with external services without appropriate assessment.
Before piloting, agree on measures of correctness, correction effort, staff usefulness and operating cost. A convincing demo alone is not sufficient evidence.
Does the 2030 strategy automatically provide funding?
No. A strategy report is not an individual funding award. Actual programmes have their own eligibility rules, deadlines and budgets that must be verified with the responsible authorities.
Define a viable use case first, then consider potential funding separately.
What is a sensible path toward 2030?
Build staff skills, improve data quality, document workflows and prioritise manageable tasks. These investments are useful even before a large AI project begins.
A small, measurable pilot can inform whether wider integration is worthwhile. The decision should follow business outcomes, not pressure to reach an abstract adoption figure.
A national strategy is not a business case
The AI Nation 2030 agenda increases attention on practical AI. For each business, the relevant question remains whether a solution improves real work and can be operated responsibly. WF Future can help assess engineering feasibility and integration requirements.