As we enter the summer of 2026, the image of a tax inspector bursting into a shop with a notepad is officially a relic of the past. In its place, a new generation of 'digital Rambos'—armed not with weapons, but with machine learning algorithms and access to global databases—is scanning economic activity in real-time. The Independent Authority for Public Revenue (AADE) has now fully deployed 'Elenxis 2.0', an AI platform that promises to plug the tax evasion leaks that have plagued Greek public coffers for decades.
The 'Wealth' Algorithm and Data Cross-Referencing
The heart of the new strategy lies in AI's ability to process vast amounts of data from disparate sources. The system is no longer limited to bank transactions and tax returns. Using neural networks, AADE is moving towards 'indirect audit methods' that analyze the lifestyle of taxpayers. From private school fees and foreign card spending to electricity consumption and social media posts, nothing goes unnoticed. If an individual's digital footprint suggests a luxurious lifestyle that is inconsistent with their declared income, the algorithm triggers a red flag.
The 'usual suspects'—freelancers with low declared incomes, short-term rental owners, and catering businesses—are at the center of the crosshairs. The integration of POS systems with cash registers, completed last year, now provides the necessary data to train AADE's predictive models. The algorithm can now identify 'unusual behaviors,' such as a restaurant issuing receipts only during off-peak hours, or a professional consistently declaring income just below the VAT threshold.
The Ethics of Surveillance and the Right to Defense
However, the use of AI in tax administration raises serious questions about transparency and civil rights. The concept of the 'black box' in AI is ever-present: how can a taxpayer defend themselves when the audit decision is based on an opaque algorithm? Legal circles warn of the risk of 'algorithmic bias,' where specific social groups or professions are disproportionately targeted due to flawed correlations in the training data.
AADE responds that the final decision to impose a fine remains with the human inspector. AI acts as a 'filter' that prioritizes cases with the highest probability of non-compliance, saving thousands of man-hours. Nevertheless, the pressure to meet revenue targets set under the Recovery and Resilience Fund is immense, and the fear of 'automated injustice' remains palpable.
Global Dimensions and Information Exchange
Greece is not acting alone in this field. The new European directive DAC8 mandates the automatic exchange of information on transactions in cryptocurrencies and digital assets, closing one of the last havens for undeclared wealth. Greek AI is now interconnected with the systems of other EU member states, making the detection of offshore companies and bank accounts in tax havens an almost automated process. The era when tax evasion was a low-risk national 'sport' seems to be ending, as technology makes the cost of hiding income higher than the cost of compliance.
- Automated lifestyle analysis through Big Data and social media.
- POS-Cash register synchronization for real-time discrepancy detection.
- Implementation of the DAC8 directive for cryptocurrency monitoring.
- Use of predictive analytics to prioritize tax audits.
In conclusion, the transition to algorithmic tax administration is an irreversible process. The challenge for the Greek state is to ensure that this powerful technology is used to establish tax justice and not as a tool for arbitrary pressure on consistent but financially vulnerable citizens. Algorithmic transparency and personal data protection will be the next major stakes in the relationship between the state and the citizen.