The submission of the new Artificial Intelligence (AI) bill to the Hellenic Parliament marks a historic turning point for the country's digital governance. This is not merely a formal transposition of European directives, but a concerted effort by Greece to define the rules of engagement in a field that is radically transforming the economy, public administration, and social cohesion. Following the European AI Act, the bill seeks to strike a delicate balance between fostering innovation and safeguarding the fundamental rights of citizens.
Establishing the National AI Supervisory Authority
A central pillar of the new legislative initiative is the creation of a National Supervisory Authority. Its role is envisioned as not just punitive or regulatory, but primarily advisory. This Authority will be responsible for certifying high-risk AI systems, ensuring that algorithms used in critical sectors—such as healthcare, justice, and labor—operate with transparency and without bias. The challenge here is twofold: firstly, staffing the body with specialized scientific personnel capable of 'reading' the code, and secondly, avoiding the creation of a new bureaucratic layer that could stifle startups.
"Artificial Intelligence is no longer a future scenario, but the daily reality of Greek administration. We must harness it before it outpaces us," the explanatory memorandum notes.
AI in the Public Sector: Efficiency or Surveillance?
The bill places significant emphasis on the use of AI systems by the state. From automated tax return processing to the deployment of chatbots for citizen services, the goal is a drastic reduction in bureaucracy. However, the use of algorithms in decision-making processes that affect citizens' lives raises serious concerns. The bill mandates a "Fundamental Rights Impact Assessment" for every new system introduced in the public sector. This means the state must prove beforehand that the algorithm does not discriminate based on gender, origin, or economic status.
- Establishing rules for the use of Generative AI in official government documentation.
- Creation of 'regulatory sandboxes' to test new technologies in a controlled environment.
- Strict penalties for the use of social scoring systems, which are explicitly prohibited.
Ethics and Transparency: The Data Challenge
One of the most thorny issues addressed by the bill is the provenance of data used to train AI models. Greece, following the European line, now demands full transparency regarding data copyrights. Companies developing models within Greek territory must publish summaries of the content used for their training. This represents a victory for creators and journalists who see their work being 'consumed' by machines without compensation. Furthermore, the bill introduces the concept of 'explainability,' requiring systems to be able to explain to the user how they arrived at a specific decision or recommendation.
Investment and Innovation: The Growth Gambit
Beyond restrictions, the bill aspires to make Greece an attractive destination for AI investment. The creation of regulatory sandboxes allows companies to develop innovative products without the fear of immediate sanctions, under the Authority's supervision. This is considered vital for the domestic startup scene, which often faces legal uncertainty. The question remains whether the Greek market can compete with tech giants from the US and China, or if it will be limited to the role of a regulator for technology that is entirely imported. The success of the bill will be judged in practice: by whether it manages to protect the citizen without driving away talent and capital.
