BgGPT and BRAIN++ | Bulgaria in the AI Age: Between Digital Transformation and Structural Contradictions – Small Country, Great Potential – Image: Xpert.Digital
Paradoxical Bulgaria: Hardly any digitalization in everyday life, but world-class in artificial intelligence
The EU AI Act: A multi-billion dollar market: How Bulgaria could become Europe's most important AI hub
The billion-dollar AI business: Why Bulgaria wants to become Europe's new Silicon Valley
Bulgaria and Artificial Intelligence? Anyone who thinks of Paris, Munich, or London as Europe's technology leaders is currently overlooking one of the continent's most fascinating markets. At first glance, the country lags significantly behind the EU average in fundamental digitalization. But away from the mainstream, a quiet sensation is unfolding in Sofia: With globally networked elite institutes, its own open-source language models, and the construction of one of Europe's first "AI factories," Bulgaria is positioning itself as a strategic bridgehead in the European tech ecosystem. Especially in times of the stringent EU AI Act and the growing demand for data-sovereign solutions, the country offers an almost unbeatable combination: highly qualified IT talent and Western European legal certainty at a fraction of the usual development costs. A look at a country that, in the age of AI, is facing the greatest economic opportunity in its recent history.
Related to this:
Small country, big potential – why Bulgaria must take the AI gamble to avoid being left behind
The underestimated AI hotspot: Why European companies now need to look to Sofia
Bulgaria stands at a historic crossroads in its economic policy. The question of whether the country will seize the opportunities of artificial intelligence or once again become a passive observer is not merely a technological one – it is a question of economic survival in the 21st-century European single market. AI development in Bulgaria presents a complex picture: surprising strengths and chronic structural weaknesses coexist, and this very tension makes the country one of the most interesting areas to observe within the European Union.
Where Bulgaria stands today: A sobering assessment
Looking at the raw figures, a sobering picture emerges. In 2024, only 6.5 percent of Bulgarian companies with ten or more employees had integrated AI technologies into their operations – a figure that places Bulgaria 25th among the 27 EU member states. By comparison, the EU average in the same year was 13.5 percent, while leading countries like Denmark and Sweden recorded adoption rates of 27.6 and 25.1 percent, respectively. Only Poland and Romania lagged behind Bulgaria. This is not merely a snapshot, but rather an expression of deeper structural problems: a lack of widespread basic digital skills, a pronounced gap between urban metropolitan centers like Sofia and rural regions, and a business landscape traditionally dominated by small and medium-sized enterprises (SMEs) that have limited resources for digital transformation.
The lag becomes even more apparent when considering the overall adoption of advanced digital technologies. Only 29.3 percent of Bulgarian companies use advanced technologies such as cloud services, data analytics, or AI – compared to an EU average of 54.6 percent. These figures suggest that this is not a deliberate reluctance specific to AI, but rather a general delay in the digital penetration of the economy. However, it would be a mistake to conclude that Bulgaria is irrelevant in the AI age. Beyond the broad adoption figures for the economy, a surprisingly dynamic AI scene is developing in a few, but highly relevant, sectors.
The paradox of cutting-edge research: World-class on a small scale
Perhaps the most remarkable phenomenon in Bulgaria's AI landscape is the existence of an institution that has garnered international attention: INSAIT, the Institute for Computer Science, Artificial Intelligence and Technology, founded in Sofia in April 2022 as a joint project with ETH Zurich and the École Polytechnique Fédérale de Lausanne (EPFL). Two of the world's most prestigious technical universities have joined forces with a Bulgarian institution – an unprecedented event in the country's history. The institute focuses on machine learning, natural language processing, computer vision, information security, quantum computing, and other emerging fields. It receives financial and technical support from global technology companies such as Google, Amazon Web Services, and DeepMind, and sees itself as a bridge between scientific excellence and entrepreneurial application.
INSAIT provided the first tangible proof of this structure's capabilities in November 2024: Bulgaria became the first EU member state to possess a highly developed AI model in its own language. The open-source language model, known as BgGPT, was developed by INSAIT and made publicly available on November 23, 2024. Its applications include educational content, personalized learning, legal research, and administrative support – a broad base that makes the model a national digital tool with societal reach. The fact that Bulgaria was faster than countries like Portugal, Hungary, or even Austria demonstrates that technological leadership is not necessarily tied to economic size.
BRAIN++: The quantum leap in AI infrastructure
In March 2025, Bulgaria received arguably the most significant economic boost in its recent digital history: The European High Performance Computing (EuroHPC) Joint Undertaking decided to locate one of six new European AI factories in Sofia. Bulgaria prevailed in the competition against considerably larger and economically stronger European nations – a success that can be attributed to the combination of its existing supercomputing infrastructure (Bulgaria hosts the EuroHPC supercomputer "Discoverer") and the institutional strength of INSAIT. The project, named BRAIN++, is designed with €90 million in EU funding, with the Bulgarian government committing to cover half of the costs itself from 2026 onwards.
The BRAIN++ project is conceived as far more than a mere data center. It aims to establish Bulgaria's first comprehensive AI factory as a national hub for the development, adaptation, and deployment of fundamental AI models, thereby contributing to European digital sovereignty. Specifically, four specialized fundamental models will be created: a language model with 175 billion parameters for the Bulgarian language (BgGPT), industrial robotics models for manufacturing and logistics (RoboticsBG), visual language models based on Earth observation data for precision agriculture and environmental monitoring (FORSE), and biomedical models for healthcare (MEDBG). The AI factory is designed to function as an open ecosystem: almost any company will be able to develop customized, niche-specific AI models for its own business processes there – without having to acquire its own high-performance computing resources. This could be the missing link that finally integrates the breadth of the Bulgarian economy into the AI transformation.
The regulatory foundation: The EU AI Act and its impact on Bulgaria
No AI market in Europe can be described today without considering the regulatory framework of the European Union. The EU AI Act, adopted in June 2024 and officially entering into force on August 1, 2024, is the world's first comprehensive legal framework for artificial intelligence and is directly applicable law in all EU member states – including Bulgaria. The law follows a risk-based approach: AI systems are categorized into four groups according to their potential for harm – unacceptable risk (prohibited), high risk (strictly regulated), limited risk (transparency obligations), and minimal risk (virtually unregulated). The most serious violations, such as breaches of the prohibition of unacceptable AI practices, can be punished with fines of up to €35 million or seven percent of global annual turnover.
The law is being introduced in phases, each requiring different actions from companies. The ban on unacceptable AI practices has been in effect since February 2, 2025; the obligations for general-purpose AI (GPAI) models have been mandatory since August 2, 2025; and the comprehensive obligations for high-risk AI systems will come into force on August 2, 2026. In Bulgaria, the Ministry of Electronic Administration has assumed responsibility for coordinating national implementation, although the appointment of supervisory and notification authorities has been delayed due to political transition phases. The Data Protection and Communications Authority (CPDP) is designated as the enforcement body for practical implementation.
One aspect particularly relevant for businesses is that the EU AI Act holds not only AI developers accountable, but also so-called deployers – companies that use existing AI systems. A Bulgarian company using an off-the-shelf AI tool for applicant selection is considered a deployer of a high-risk AI system under Article 26 of the Act and is subject to specific documentation, monitoring, and transparency obligations. These requirements affect a broad cross-section of Bulgarian businesses, which have previously had little experience with structured AI governance. At the same time, the AI Act creates a historically rare market opening for competent providers of compliance solutions: Demand for risk assessment, documentation tools, and governance platforms is exploding across Europe.
Bulgaria's national AI strategy: A vision without a roadmap
Bulgaria has had a national AI strategy since 2020, comprising six main pillars: infrastructure, education, research, data potential, sectoral innovation, and ethical AI development. This strategy establishes a sound conceptual foundation in principle. However, it suffers from a serious structural deficiency: it lacks a binding action plan with concrete implementation measures and measurable deadlines. This is not unique to Bulgaria – Bulgaria was among the EU member states that missed the deadline set under the EU Coordinated Plan for AI for submitting a national AI strategy. The lack of operational planning has historically meant that strategic declarations of intent have not been translated into economically effective measures.
Nevertheless, there are signs of a realpolitik course correction. The Bulgarian government has pledged €92 million in investments from EU structural funds and national budget resources as part of its national digital transformation. The allocation of these funds is remarkably strategic: €30 million will go towards AI research infrastructure, including GPU clusters and open AI platforms; €25 million is earmarked for direct SME digitalization subsidies; €20 million will be invested in the digital transformation of the public sector; and €17 million will be provided for further education, AI/ML training programs, and vocational retraining. This allocation demonstrates a clear prioritization: not only is cutting-edge research to be promoted, but the broader economy is also to be involved in the digital transformation through targeted incentives.
The skilled worker dilemma: Talent and brain drain
Bulgaria boasts a remarkably high-performing IT community. With over 80,000 highly skilled IT professionals, a strong presence of international technology companies such as SAP Labs, HPE, Bosch, and Broadcom in Sofia Tech Park, and approximately 833 IT companies operating in the fields of outsourcing, software development, and research and development, the country has established itself as a mature and competitive IT ecosystem. Bulgarian developers are regularly ranked among the world's best on international comparison platforms; English language proficiency is comparatively high, and the cost structure remains highly attractive for Western European companies: hourly rates range from €25 to €50, representing savings of 40 to 60 percent compared to similar Western European markets.
This potential, however, is significantly limited by a structural brain drain: Bulgaria has been suffering for decades from one of the most intense brain drain dynamics in the EU. Between 1992 and 2015, an estimated three million Bulgarians left their country – a historically unprecedented exodus, given a total population of around seven million at the time. Highly skilled IT professionals and engineers are migrating to well-paid positions in Germany, the Netherlands, Austria, and the United Kingdom – countries where not only wages but also quality of life, infrastructure, and career development opportunities are perceived as more attractive. This brain drain threatens the long-term foundation of any AI ecosystem: the availability of excellent minds. INSAIT and BRAIN++ are attempts to counteract this dynamic by creating internationally competitive research and working conditions domestically – an approach that is structurally sound but takes time to produce measurable results.
The opportunities of the AI market: Where Bulgaria becomes a leverage point
Despite the structural weaknesses described, Bulgaria offers AI companies targeting the European market a strategically outstanding combination of strengths that hardly any other EU country combines in this way.
The first and most immediate advantage is cost competitiveness within the EU legal framework. Bulgaria combines IT development costs comparable to those of non-European outsourcing centers with full protection under European IP, labor, and data protection law. The GDPR, the EU AI Act, and all other EU regulations apply in Bulgaria just as they do in Germany or France – a crucial advantage over nearshore and offshore alternatives outside the EU. The time zone overlap with Western Europe is almost complete (only a one-hour difference), enabling agile collaboration without communication barriers. Furthermore, the corporate tax rate of ten percent – one of the lowest in the entire EU – makes Bulgaria an attractive location for establishing business structures.
The second strategic lever is Bulgaria's growing role as a development hub for specialized AI solutions, particularly for European SMEs. Google, in its report on the Bulgarian market, predicts that AI could contribute up to five billion euros to Bulgaria's GDP over the next decade. This estimate is conservative, considering how the BRAIN++ ecosystem aims to empower SMEs to develop and operate customized AI models without having to maintain their own high-performance infrastructure. This creates an institutional infrastructure, especially for AI companies developing Software-as-a-Service solutions for specific business applications, which reduces development costs and accelerates market access.
Our EU and German expertise in business development, sales and marketing
Our EU and German expertise in business development, sales and marketing - Image: Xpert.Digital
Industry focus areas: B2B, digitalization (from AI to XR), mechanical engineering, logistics, renewable energies and industry
More information here:
A thematic hub offering insights and expertise:
- Knowledge platform covering global and regional economies, innovation and industry-specific trends
- A collection of analyses, insights, and background information from our key areas of focus
- A place for expertise and information on current developments in business and technology
- A hub for companies seeking information on markets, digitalization, and industry innovations
Private AI as a competitive advantage: Why companies rely on data-sovereign solutions
The compliance market as a growth driver
Three business models for AI startups in Europe: On-premise, no-code, and governance
One of the most dynamic submarkets for AI companies in Europe is currently the market for AI compliance and governance solutions – a market directly driven by the EU AI Act. The scale is impressive: the global market for EU AI Act compliance solutions was estimated at US$1.7 billion in 2025 and is projected to grow to US$16.8 billion by 2034, representing a compound annual growth rate (CAGR) of 31 percent. Europe dominates this market with a 48.5 percent market share. Even considering the more conservative estimates for the narrower market of AI governance and compliance platforms – US$440 million in 2025, projected to reach US$5.84 billion by 2034 at a CAGR of 35.7 percent – the growth dynamics are extraordinary.
What is driving this growth? At its core, it is the mandatory nature of the regulation, combined with the complexity of its implementation. Companies using AI are now required to classify their systems, implement risk management systems, meet transparency requirements, register high-risk systems in EU databases, and ensure comprehensive human oversight. These requirements are not optional – and the penalties for violations are existential enough to firmly establish the issue in the boardrooms of European companies. For AI companies offering specialized compliance tooling solutions, this opens up a market that is neither cyclically fluctuating nor voluntary: it is mandated by regulation and remains structurally stable as long as the regulation is in effect.
Private AI infrastructure: The data sovereignty revolution in the enterprise sector
Alongside compliance demands, a second, structurally even more significant market is emerging in Europe: the demand for private, data-sovereign AI deployment solutions. The reason is simple: 83 percent of EU companies now prefer on-premises or controlled cloud AI solutions to meet stringent data sovereignty requirements. More than 60 percent of enterprise AI tools lack documented data processing agreements with third-party providers. In an environment where GDPR fines can reach up to €20 million or four percent of global annual turnover, and the EU AI Act introduces comparable sanction levels, the risk of uncontrolled data flows to external AI providers is real and calculable for companies.
The market's response is the concept of private AI: AI models and infrastructures that run on the company's own or private cloud infrastructure, without sensitive data leaving the corporate environment. Deloitte explicitly describes this trend as a strategic priority for European companies: Private AI eliminates the risk of outsourcing proprietary processes, ensures compliance with GDPR and AI Act requirements, and reduces operational costs by eliminating API fees and data transfer costs. AI companies offering private, self-hosted, or on-premises solutions are thus addressing the critical juncture where European companies are caught between the pressure to innovate and the obligation to comply.
Knowledge management and enterprise AI: The mass market of enterprise applications
The third major market segment relevant for specialized AI companies in Europe encompasses AI-powered knowledge management and enterprise applications. The global market for AI-driven knowledge management is projected to reach $7.66 billion in 2025 and grow to $51.36 billion by 2030 – an annual growth rate of 46.2 percent. By 2026, 80 percent of all companies worldwide will have deployed generative AI, compared to less than five percent in 2023. Corporate spending on generative AI is expected to triple from $11.5 billion in 2024 to $37 billion in 2025 – a threefold increase in a single year.
However, this market is not homogeneous. Demand is increasingly focused on secure, GDPR-compliant enterprise applications that link existing knowledge databases, document archives, and process data with AI without requiring the transfer of sensitive information to external servers. Companies are looking for solutions that enable natural language interaction with their internal knowledge systems while maintaining complete control over data and models. This creates enormous demand for AI providers who can offer precisely this combination of user-friendliness and data sovereignty, a demand that none of the major US hyperscaler offerings fully meet—not least because these offerings structurally include non-European data paths.
In parallel, the market for agentic AI – AI systems capable of independently executing multi-stage workflows, making decisions, and interacting with other systems – is developing into one of the most dynamic growth segments in the European enterprise sector. The European market for enterprise agentic AI was valued at US$634 million in 2024 and is projected to grow to US$5.6 billion by 2030, representing an annual growth rate of 44.5 percent. This growth is driven by regulatory pressure from the EU AI Act, which, coupled with a structural shortage of qualified technical personnel, is making the automation of cognitive workflows a business necessity, not merely an efficiency improvement.
The strategic positioning of AI providers in the Bulgarian context
For AI companies that either operate from Bulgaria or want to use the Bulgarian and Eastern European market as a springboard into the overall European market, the described market situation offers several strategic positioning options, each based on real and growing market needs.
The first path is to focus on data-sovereign enterprise AI platforms. European companies – and especially those in the DACH region, France, and the Nordic countries – are actively seeking AI solutions that can be operated entirely within their own IT infrastructure, without relying on external model providers. This creates a structural market gap that is particularly relevant for AI providers offering a self-hosted or on-premises architecture. The combination of a small operational footprint, European legal certainty, and the ability to work with models from different providers presents a compelling value proposition for medium-sized and large companies – especially in a market environment where vendor lock-in for AI infrastructure is perceived as a growing strategic risk.
The second path leads through AI-powered knowledge and information management for companies. In a world where employees are confronted daily with exponentially growing amounts of information, and where, at the same time, the requirements for documenting decisions and processes are increasing due to regulatory frameworks such as the AI Act, the need for intelligent systems for structuring, finding, and utilizing the company's internal knowledge base is fundamental. Solutions that enable secure, GDPR-compliant, AI-powered searching and interaction with internal documents appeal to a broad target group including legal departments, compliance teams, research and development departments, and operational management.
A third path opens up via the no-code or low-code approach to AI deployment. Given that the majority of the European economy consists of small and medium-sized enterprises (SMEs) that lack dedicated data science teams and substantial IT budgets, the market for platforms enabling AI integration without programming skills is structurally attractive. Bulgaria—with its BRAIN++ ecosystem and the infrastructure of Sofia Tech Park—offers an increasingly favorable environment for developing such solutions, characterized by low development costs, a growing regulatory expertise base, and readily available academic competence.
Risks and structural limitations
A balanced analysis cannot ignore the significant risks and limitations that restrict Bulgaria's AI ambitions. The most serious structural problem is, and remains, the brain drain. Even if INSAIT and BRAIN++ attract talent to the country in the long term, the current dynamic is still predominantly driven by emigration. While average salaries in the Bulgarian IT sector are competitive by regional standards, they are far below the salary levels that qualified AI engineers can achieve in Munich, Vienna, or Amsterdam. The cost-of-living gap mitigates this difference, but does not eliminate it entirely—especially when career opportunities, networking access, and international visibility are also factored into the decision-making process.
The second risk is institutional in nature: Delays in the appointment of national supervisory authorities for the EU AI Act, the lack of concrete implementation measures for the national AI strategy, and the political instability that has repeatedly forced Bulgaria into early elections in recent years create an environment that makes long-term strategic planning difficult for companies. Regulatory predictability is a key prerequisite for investments in AI infrastructure and development – and this predictability is not yet fully present in Bulgaria.
A third risk arises from market size: with a population of approximately 6.5 million, Bulgaria's domestic market is not large enough to be attractive to globally oriented AI companies solely as a sales market. Bulgaria's value lies not primarily in the local market, but in its function as a cost-effective, EU-compliant development location and as a gateway to the pan-European market. AI companies operating in or from Bulgaria must therefore focus their commercial efforts on the pan-European market.
An outlook: Bulgaria between digital catching up and strategic leadership
The simultaneous presence of lagging behind and excelling, institutional weakness and scientific excellence, brain drain and infrastructure investment makes Bulgaria's AI development one of the most fascinating case studies in EU digital policy. Over the past three years, the country has initiated developments that are considerably bolder and more forward-looking than its broad economic starting point would suggest. INSAIT, BgGPT, BRAIN++, the European AI Factory in Sofia – these names represent a strategic shift whose economic benefits have yet to fully materialize, but whose direction is clearly discernible.
For AI companies looking to expand their market position in Europe or seeking an EU-compliant development location, Bulgaria in 2026 offers a set of advantages that is virtually impossible to replicate: a combination of low development costs, high technical expertise, full EU legal certainty, emerging world-class AI infrastructure, and a business landscape hungry for digitalization support. The decisive competitive advantage lies not in the size of the local market, but in Bulgaria's unique positioning as an affordable, legally secure gateway to the world's largest economic area – the European Single Market with 450 million consumers and an AI compliance market that is projected to grow to over US$16 billion by 2034.
The real economic question for Bulgaria is ultimately not whether the country can build AI capabilities – it can, as recent developments prove. The crucial question is whether the political class, the business landscape, and the scientific community act quickly and in a coordinated manner enough to seize this window of opportunity before it closes. Market forces wait, and in a sector where technological advantages emerge and disappear within months, speed is just as important as strategy.
🎯🎯🎯 Data-driven B2B industry hub as a quasi-in-house solution
The quasi-in-house solution: How Xpert.Digital closes operational gaps in B2B marketing and sales – Smart Content-Driven Business - Image: Xpert.Digital
Xpert.Digital is a data-driven B2B industry hub led by Konrad Wolfenstein . The company acts as an external, quasi-in-house solution for industrial partners, closing operational gaps in marketing, content, and sales – without requiring additional resources on the client side.
More information here:
Your global marketing and business development partner
☑️ Our business language is English or German
☑️ NEW: Correspondence in your native language!
Konrad Wolfenstein
I and my team are happy to be available to you as your personal advisor.
You can contact me by filling out the contact form here wolfenstein@xpert.digital:or simply call me at +49 7348 4088 965. My email address is
I'm looking forward to our joint project.
