Research
Greece
Contextualization
The discussions in Greece about automation have become intense both at the institutional and the public level. According to the Digital Government factsheet for Greece (European Commission, 2019), the Greek government – in cooperation with the Ministry of Digital Policy, Telecommunications and Information – aims to drastically transform the Hellenic State. This involves simplifying procedures and implementing big changes, especially in the field of new technology.
Moreover, with its National Digital Strategy (Hellenic Ministry of Digital Policy, Telecommunications & Media, 2016), Greece aims to kick-start its economic development, create new jobs, enhance the efficiency of public administration, eliminate geographical and social exclusion, and strengthen the inclusion of all citizens in the economic and social life of the country. This strategy focuses on seven priorities, with an emphasis on the digitalization of public governance and the strengthening of security.
The Greek government has not yet adopted a national AI strategy. However, ADM systems are thriving at the national level, especially on research projects, and there are a considerable amount of use-cases. Academic institutions and scientific research centers in Greece focus their activities on the development and deployment of ADM tools based on national grants, funds from the European Union (such as the Horizon 2020 programs), and collaboration with private industry. Finally, the Greek state is struggling to keep up with the technological developments in ADM and reflect on the challenges that arise by participating in common initiatives at the EU and Council of Europe level, and by establishing scientific committees for the examination of the role of algorithms and AI.
A catalog of ADM cases
Patrolling and surveillance at the borders
The Center for Research and Technology Hellas (ITI/CERTH), the National and Kapodistrian University of Athens (UoA), and the Hellenic Ministry of Defense participate in the research project consortium ROBORDER (“autonomous swarm of heterogeneous RObots for BORDER surveillance”). The aim of the ROBORDER project is to deliver a fully-functional, autonomous border surveillance system. The system consists of unmanned mobile robots including aerial, water surface, underwater and ground vehicles, capable of functioning both on a standalone basis and in swarms. The project started in May 2017 and will conclude in February 2021. Twelve other countries are also participating in this consortium: Portugal, United Kingdom, Romania, Germany, Italy, Switzerland, Bulgaria, Spain, Hungary, Belgium, Finland, and Estonia (ROBORDER, 2017).
Furthermore, a considerable number of public and private stakeholders in Greece are taking part in a research project called TRESSPASS (“robusT Risk basEd Screening and alert System for PASSengers and luggage”). The Greek National Center for Scientific Research “Demokritos” is coordinating the project, while six other Greek actors are part of the project’s consortium. These include the Piraeus Port Authority, the Greek Ministry of Maritime Affairs and Insular Policy, the Hellenic Customs Administration, and the Center for Security Studies (KEMEA), which is a research center within the Hellenic Ministry of Citizen Protection. TRESSPASS promotes the idea of “risk-based” security checks and offers an analytic framework for modelling risk as well as a systematic approach of quantifying risk, based on a set of indicators. Once an estimate of these indicators is made using data collected from background information, sensors and applications, a level of risk is calculated for each traveler, based on the risk management approach introduced in the TRESSPASS project (TRESSPASS, 2018).
In the field of border management, another research project that has attracted attention over the past year is the iBorderCtrl project (“Intelligent Portable Control System”). According to the project’s website, the aim of this project is to enable faster and more thorough border control for third-country nationals crossing the land borders of EU Member States. iBorderCtrl was implemented on the Hungarian, Greek, and Latvian borders as a pilot project, but it is not an authorized law enforcement system. KEMEA and the Train- OSE S.A., a Greek railway company, participated in piloting the project in Greece. The project came to an end in August 2019 (iBorderCtrl, 2016). In November 2018, the Greek civil society organization Homo Digitalis filed a petition to the Greek Parliament regarding the pilot implementation of the iBorderCtrl project at Greek borders (Homo Digitalis, 2018). Finally, based on a successful access to information request it was revealed that during the pilot project in Greece the role of travelers was primarily “played” both by the Border Guards and Hellenic Police Officers. Furthermore, staff from KEMEA and other consortium members voluntarily played the role of travelers. Thus, it appears that no real travelers participated in the Greek pilots (iBorderCtrl, 2019).
Finally, KEMEA is also involved in another research project – related to the use of ADM for border management – known as FOLDOUT (“Through-foliage detection, including in the outermost regions of the EU”). The aim of FOLDOUT is to build a system which combines various sensors and technologies and fuses them into an intelligent detection platform for border surveillance. According to FOLDOUT’s website, the project will improve the speed of the work of border guards by combining events from various sensors to give a situation threat assessment followed by suggested reaction scenarios. According to the project’s website, a two-year pilot project in Bulgaria and demonstrations of the project in Greece, Finland, Lithuania, and French Guiana are all planned. However, there are no details on the exact dates of such demonstrations. Nevertheless, the goals are described as follows: detection of irregular border crossings (illegal migrants + vehicles) in forest terrain border surveillance for the Bulgarian and Greek demonstrations, and detection of illegal transport and entry of goods (trafficking) in temperate broadleaf forest and mixed terrain border surveillance for the demonstrations in Finland, Lithuania, and French Guiana (FOLDOUT, 2018).
Automated car insurance
In November 2017, two companies – Interamerican S.A. and OSEVEN – informed the Hellenic Data Protection Authority (HDPA) about their car insurance program called “Smart Drive”. The price of the insurance for the “Smart Drive” program is determined by the driver’s behavior based on an application installed in the driver’s smartphone. Following two years of wide consultations between the HDPA and the companies, the HDPA announced they would postpone the adoption of a decision on this case until the European Data Protection Board has finalized its guidelines on processing personal data in the context of connected vehicles and mobility related applications (Hellenic Data Protection Authority, 2020a). The related decision that the HDPA will hand down in the coming months is awaited with much interest.
Autonomous driving
In 2015, the city of Trikala was in the spotlight because of a self-driving, 10-passenger electric bus. The autonomous vehicle was part of a European Union-funded project called CityMobil2 (CM2). The maximum speed of the bus was only 20 kilometers per hour, and it could not change lanes, however it stopped when any obstacles, such as parked cars, got in its way (Rainey, 2018). Now, the city is looking into autonomous driving a little further. AVINT (“Autonomous vehicles integrated within the urban context”) is a project set up to study urban transport in Trikala. It will integrate a line of automated buses – three buses in total, carrying 10 passengers each – into the city’s transportation network. This automated bus line will connect Trikala city center with the university campus. The project started in January 2019 and will be completed in December 2021. According to the project’s website, Greece is a pioneer in autonomous driving, as it has had a legal framework in place since 2014. The framework allows for the operation of automated buses without a driver under strict conditions. The partners in the project are the e-Trikala company, owned by the municipality of Trikala, the Institute of Communications and Computer Systems (ICCS) of the School of Electrical and Computer Engineering at the National Technical University of Athens, and the private company Space Hellas (AVINT, 2019).
Lastly, the University of Patras, ITI/CERTH, and the Foundation for Research and Technology Hellas (FORTH) took part in the ADAS&ME project (“Adaptive ADAS to support incapacitated drivers & Mitigate Effectively risks through tailor-made HMI under automation”). The aim of the research project is to develop robust detection/prediction algorithms to monitor the physical and emotional state of the driver/rider to mitigate accidents related to driver fatigue, sleepiness, stress, inattention or high emotions. The project focused on seven use-cases, covering a wide variety of critical driving scenarios, using different types of vehicles (cars, trucks, buses, and motorcycles). The results of the project were demonstrated in December 2019 and the project came to an end in February 2020 (ADAS&ME, 2016).
Healthcare and medicine
NCSR Demokritos is the coordinator of the research project IASIS (“Integration and analysis of heterogeneous big data for precision medicine and suggested treatments for different types of patients”). In addition, the private Greek company Athens Technology Center also participates in this project. The goal of IASIS is to integrate data from various sources, including genomics, electronic health records and bibliography, and apply advanced analytical methods to discover useful patterns. Data resources for two different disease categories, namely dementia, and lung cancer, will be explored during the project’s implementation phase. It is not known whether and how the consortium aims to integrate such a system into clinical decision making (IASIS, 2017).
Furthermore, Aristotle University of Thessaloniki (AUTH) coordinated the research project i-PROGNOSIS (“Intelligent Parkinson eaRly detectiOn Guiding NOvel Supportive InterventionS”), while three other Greek stakeholders, including ITI/CERTH, the Greek telecommunications company COSMOTE, and the Microsoft Innovation Center – Greece, participated in it. Among the aims of the project was a plan to develop and employ advanced big data analytics and machine learning techniques, in a distributed and privacy-aware fashion, to create a Parkinson’s Disease Behavioral Model and construct reliable early symptoms detection alarms. The project concluded in January 2020 (i-PROGNOSIS, 2016).
Finally, the pharmaceutical company Pfizer has chosen the city of Thessaloniki for the creation of one of its six digital laboratories, aimed at capitalizing on technologies such as AI and big data analytics for the development of new treatments. The investment was announced in September 2019 by the Greek Prime Minister, Mr. Kyriakos Mitsotakis, at the 84th edition of the Thessaloniki International Fair (TIF), the largest exhibition event in the Balkans (Konti, 2019).
Automated processing of traffic offenses and Automatic Number plate recognition cameras
Automated processing of traffic offenses was introduced in Greece back in 2001 based on Presidential Decree 287/2001. However, in recent years, because of the financial crisis, the purchase of photo radars has been limited. As a result, the number of available traffic enforcement cameras on national highways is low. According to a 2017 interview with the then Deputy Minister for Citizen Protection, Mr. Nikolaos Toskas, traffic enforcement cameras were only present on a single section of the entire national highway network. However, 14 traffic enforcement cameras have been installed on the Attica toll road, which is a privately owned motorway system (Kathimerini, 2017). It is also worth noting that the number of normal traffic cameras on national highways is high, while police officers use manually operated tools to detect drivers who exceed the speed limit (Lialos, 2017).
In addition, Automatic Number Plate Recognition (ANPR) cameras were openly introduced in 2019. According to a Greek news report, around 50 Hellenic police vehicles are equipped with such cameras and they are capable of detecting the license plates of stolen vehicles or vehicles used in criminal activities. The ANPR can recognize 15,000 license plates per hour, and the cameras can detect license plates at speeds above 170 km per hour. In the near future, they will also be able to detect uninsured vehicles (To Vima, 2019).
Predictive policing
Greece participates in several predictive policing projects. Firstly, in the spring of 2019, Intracom Telecom, a global telecommunication systems and solutions vendor, signed a four million euro contract with the Hellenic Police and the Hellenic Ministry of Citizen Protection for the development of “Smart Policing” software. The contract states that Intracom Telecom will develop an integrated information system which will include the core building blocks for face recognition, automated fingerprint identification, document authentication, and complex information searches in existing databases. The system will allow the police to run checks in real-time during police stops (Intracom Telecom, 2019 and Hellenic Police, 2019). Based on the implementation timetable of the project, the vendor will deliver the system to the Hellenic Police within 20 months of signing the contract (Hellenic Police 2018).
Secondly, the Hellenic Police and four other law enforcement authorities (i.e. the West Midlands Police (UK), the Police and Crime Commissioner for Thames Valley (UK), the Serbian Ministry of Interior (RS), and the Police Academy in Szcytno (PL)) are taking part in a project called SPIRIT (“Scalable privacy-preserving intelligence analysis for resolving identities”). The information available on the project’s web- site is very limited, even though the project has been up and running since August 2018. Nevertheless, it is evident from the website that the project aims to use tools such as face extraction and matching, to correlate information from social media data, and to continuously initiate complex associative searches across all sources relevant to criminal investigation. A successful freedom of information request revealed that trial runs will be organized, including in Greece, involving end-users and practitioners, such as law enforcement agencies, and other stakeholders. The Hellenic Police (Border Unit) is involved in trials that run between January and August 2020, while reports of trial runs and tests are expected in the form of deliverables in April 2020 and January 2021 (Ask the EU, 2020). Two of the Greek stakeholders in SPIRIT’s consortium are the private companies Singular Logic and Nydor System Technologies. The project is due to end in July 2021 (SPIRIT, 2018).
Thirdly, the Center for Security Studies (KEMEA) is a member of the COPKIT project, which is developing data-driven policing technologies to support law enforcement agencies (LEAs) and help prevent, investigate and mitigate crime and terrorism. Europol and nine LEAs from eight different countries are members of this project as well. The first private demo of the COPKIT tools took place in Athens in November 2019 at the premises of KEMEA. A second private demo was held in Madrid in May 2020 and the third will take place in Brussels in November 2020, followed by a public demonstration at the project’s final conference (COPKIT, 2018).
Furthermore, ITI/CERTH is a participant in the ANITA research project (“Advanced tools for fighting oNline Illegal TrAfficking”), which started in May 2018 and is set to conclude in April 2021. The aim of the project is to design and develop a knowledge-based user-centered investigation system for analyzing heterogeneous (text, audio, video, image) online and offline content for fighting the trafficking of drugs, counterfeit medicines, and firearms. According to the project’s website (ANITA 2018), ANITA will combine the following benefits:
- data source analysis of crypto-currency networks, transactions, and blockchain technologies
- advanced big data analytics tools for the automatic analysis of vast amounts of multimodal content from identified sources
- methodologies for capturing, modelling and inferring knowledge in human-understandable forms (e.g. expressive ontologies), also extracting relevant and new knowledge from neural networks and formally storing it in the form of
- domain-related and user-oriented intelligence applications, which will enable users to identify patterns for correlations among illegal trafficking events, entities and activities, and will support decision- making processes for
- lastly, the project’s goal is to develop a cognitive user modelling framework that will capture, analyze, interpret, mimic, and integrate key human cognitive and information processing functions to represent the incorporation of human perception/cognition principles in the system processing pipelines (i.e.integrating the investigators/officers/practitioners ‘in-the-loop’ of the overall analysis process).
Finally, ITI/CERTH also participates in a project called CONNEXIONs (“InterCONnected NEXt-Generation Immersive IoT Platform of Crime and Terrorism DetectiON, PredictiON, InvestigatiON, and PreventiON Services”), which started in 2018 and will end in August 2021. The aim of the project is to enhance operational and (near) real-time situational awareness of the consortium’s law enforcement authorities, through automated identification, interpretation, fusion and correlation of multiple heterogeneous big data sources. Such multimodal data includes surface, deep, and dark web, as well as social media content in seven languages (English, French, German, Portuguese, Romanian, Spanish, and Arabic). Other sources include data acquired by IoT devices and digital evidence (CONNEXIONs, 2018).
Policy, oversight and public debate
Policy
Establishment of the Legislative Committee for the transposition of the new Copyright Directive and its controversial.
Article 17
In September 2019, Ms. Lina Mendoni, Greek Minister of Culture and Sports, issued her decision to establish a legislative committee entrusted with the transposition of Directive 2019/790 on copyright and related rights. There is great interest in the transposition of the controversial Article 17, which relates to the use of content moderation software in the Greek legal framework. The Committee is scheduled to deliver its work by the end of June 2020 (Hellenic Ministry of Culture and Sports, 2019).
Greece at the Ad Hoc Committee on AI of the Council of Europe
In September 2019, the Committee of Ministers of the Council of Europe set up an Ad Hoc Committee on AI (CAHAI). The Greek delegation at the CAHAI is composed of representatives from the Ministry of Foreign Affairs and the Ministry of Digital Governance. The Greek digital rights organization Homo Digitalis is an observer of the CAHAI since June 2020, as well. The Committee aims to examine the feasibility and potential elements of a legal framework for the development, design and application of AI, based on the Council of Europe’s standards on human rights, democracy and the rule of law (Council of Europe, 2019a).
Greece finally enforces the GDPR and Law Enforcement Directive, but raises criticism
In August 2019, the Greek Parliament adopted the Law 4624/2019, which enforces the provisions of the GDPR and implements the Directive 2016/680 (the Law Enforcement Directive – LED) into the national legal framework. Article 52 of L.4624/2019 transposes Article 11 of the LED (Automated individual decision-making). In September 2019, the Greek digital rights organization Homo Digitalis and the Greek consumer rights organization EKPIZO challenged the clarity of L.4624/2019 and requested the Hellenic Data Protection Authority to issue an Opinion on its conformity with the provisions of the GDPR and the LED (Homo Digitalis & EKPIZO, 2019). Also, in October 2019, Homo Digitalis filed a complaint (CHAP(2019)03059) to the European Commission against a breach of the provisions of the EU data protection law (GDPR and LED) by Greece and L.4624/2019. According to Homo Digitalis, while Article 52, para.1 of the L.4624/2019 states that a decision based solely on automated processing, including profiling, which produces an adverse legal effect concerning the data subject or significantly affects him or her is only allowed when it is authorized by law. It does not state that such a law shall provide appropriate safeguards for the rights and freedoms of the data subject, not even the right to obtain human intervention on the part of the controller. This omission is very serious and could jeopardize the rights of the data subjects in the age of technology-led predictive policing (Homo Digitalis, 2019a). The Hellenic DPA issued its Opinion on L. 4624/2019 in January 2020, which heavily criticized the lack of conformity of some of its provisions with the GDPR and the LED. Concerning Article 52, the Hellenic DPA stated that Article 11 LED has been poorly transposed because the L. 4624/2019 does not provide the appropriate safeguards for the rights and freedoms of the data subject, at least the right to obtain human intervention on the part of the controller (Hellenic Data Protection Authority, 2020b, p.23).
Scientific Committee to examine the impact of AI in the Greek Judicial System
In February 2019, Mr. Michael Kalogirou, the then Minister of Justice, Transparency, and Human Rights, issued a Ministerial Decision to establish a scientific committee entrusted with the task of examining the impact of AI on the Greek judicial system. The scientific committee is composed of members of the Greek Courts and Supreme Courts, academics, members of Greek law bars, and civil servants. The committee will submit proposals, inter alia, to the ministry, for the modernization of the existing legal framework taking into account technological developments (Hellenic Ministry of Justice, Transparency, and Human Rights, 2019).
The potential impact of AI on human rights in the Constitution?
In February 2019, during the Greek Parliament’s Plenary Session on the revision of the Greek Constitution, Mr. Yiannis Maniatis acting as a rapporteur for the Movement for Change (KINAL) political party, proposed an amendment to Article 25 of the Greek Constitution regarding the potential impact of AI on the rule of law, and respect for human rights and fundamental freedoms. Τhe suggestion was not voted through (Maniatis, 2019).
Greece signs a motion for the examination of the role of algorithms and AI in the criminal justice system
In September 2018, Ms. Ioanneta Kavvadia, the then representative of Greece at the Parliamentary Assembly (PACE) of the Council of Europe, signed a joint motion with more than twenty other representatives. The petition requested the PACE to develop a recommendation examining the role of algorithms and AI in the criminal justice system bearing in mind the Council of Europe standards on human rights and the rule of law (Cilevičs et al, 2018).
Greece signs a declaration of cooperation on AI with EU Member States
In April 2018, 25 EU Member States signed a declaration of cooperation on AI. With this declaration, the Member States agreed to work together on different challenges raised by AI, from ensuring Europe’s competitiveness in the research and deployment of AI, to dealing with social, economic, ethical, and legal questions. In May 2018, Mr. Costas Fotakis, the then Alternate Minister for Education, Research, and Religious Affairs, signed the declaration on behalf of Greece (Hellenic Republic, 2018).
Civil Society and academia
Hellenic Data Protection Authority (HDPA)
The Hellenic Data Protection Authority (HDPA) is a constitutionally consolidated independent authority. The HDPA posted guidance on its website stating that citizen rights should not to be subject to a decision based solely on automated processing, including profiling under Article 22 GDPR. By using simple language, the HDPA has attempted to clarify the GDPR’s provisions of Article 22 and explain its connection with Articles 13, 14, 17, 18, and 21 (Hellenic Data Protection Authority, 2019).
National Commission for Human Rights in the Hellenic Republic (GNCHR)
The Greek National Commission for Human Rights (GNCHR) is an independent advisory body to the state and specializes in human rights issues. It was established by Law 2667/1998, in accordance with the UN Paris Principles, and inaugurated on 10 January 2000.
Hellenic Artificial Intelligence Society(EETN)
The Hellenic Artificial Intelligence Society (EETN) started in 1996 and it is a non-profit scientific organization devoted to organizing and promoting AI research in Greece and abroad.
The Center for Security Studies (KEMEA)
The Center for Security Studies – KEMEA is a scientific, consulting, and research agency supervised by the Minister of Citizen Protection involved in many of the research projects described above.
The Foundation for Research and Technology – Hellas (FORTH)
The Foundation for Research and Technology – Hellas (FORTH) was founded in 1983 and is one of the largest research centers in Greece and it has a reputation for being a top-level international research institute.
The National Center of Scientific Research “Demokritos” (NCSR Demokritos)
Founded in 1961, the National Center of Scientific Research Demokritos is the largest multidisciplinary research center in Greece. Together with Ernst & Young Greece, it founded the Hellenic Artificial Intelligence Center of Excellence (NCSR Demokritos & EY, 2019).
The Center for Research and Technology Hellas (ITI/CERTH)
The Center for Research and Technology Hellas (ITI/CERTH) was founded in 2000 and it is listed among the TOP 20 EU research institutions with the highest participation in competitive research grants.
The Research and Innovation Center in Information, Communication and Knowledge Technologies “ATHENA”
The ATHENA Research and Innovation Center was founded in 2003. It is comprised of the Institute for Language and Speech Processing (ILSP), the Information Management Systems Institute (IMSI) and the Industrial Systems Institute (ISI).
The Institute of Communications and Computer Systems (ICCS)
The Institute of Communication and Computer Systems (ICCS) was founded in 1992 and belongs to the School of Electrical and Computer Engineering (ECE) at the National Technical University of Athens (NTUA).
The Hellenic Scientific Council for Information Society (HSCIS)
The Hellenic Scientific Council for Information Society (HSCIS) was established in 2001 and its main purpose is to contribute to the development of the Information Society within a privacy preserving environment.
The Hellenic Association of Data Protection & Privacy (HADPP)
The Hellenic Association of Data Protection & Privacy (HADPP) is a non-profit group of professionals with a passion to exchange, analyze, communicate and educate on the ongoing issues of data protection, privacy and security.
The Institute for Privacy Law, Data Protection and Technology (IPL)
The Institute for Privacy Law, Data Protection and Technology (IPL) was established by the European Public Law Organization and aims at advancing and promoting knowledge and policy in the fields of privacy and data protection law, technology, society, and fundamental rights.
The Open Technologies Alliance (GFOSS)
The Open Technologies Alliance (GFOSS) is a non-profit organization founded in 2008. Its main goal is to promote openness through the use and the development of open standards and open technologies in education, public administration, and business.
Consumers’ Association The Quality of Life (EKPIZO)
EKPIZO was established in 1988 to protect consumer rights and improving the quality of life. It is a non-governmental, non-profit association with more than 24,000 members. EKPIZO is a member of the European Consumer Organization (BEUC).
Center for Consumer protection (KEPKA)
KEPKA is a non-governmental, non-profit, independent consumer organization founded in 1982. KEPKA is a member of the European Consumer Organization (BEUC).
Key takeaways
The use of Automatic Number Plate Recognition (ANPR) cameras by the Hellenic Police is a practice that requires special attention, since it could pose important challenges for the protection of the rights of individuals in Greece. Since the cameras are deployed in police vehicles they could, in theory, monitor nearby pedestrians or scan details of vehicles that are not stolen. This raises important questions regarding the rights to privacy and data protection. The national law should provide thorough details and safeguards on the collection of this data, the purpose of the processing activities, and the retention period, as well as details of access to this information by competent authorities.
In addition, the four million euro contract given to Intracom Telecom to work with the Hellenic Police to develop an information system related to face recognition, automated fingerprint identification, document authentication, and complex information searches in existing databases, raises important human rights concerns. The police intend to use the software during routine stops to verify the identity of individuals. However, a national legal framework providing for the legal basis of such activities and for the appropriate safeguards for the rights and freedoms of the data subjects involved is lacking. Also, one could argue that such a measure does not meet the criterion of strict necessity that allows for the processing of biometric information provided for in Article 10 of the Law Enforcement Directive.
Moreover, it is clear that there are many Greek, or Greek stakeholder-led, research projects with a focus on the development and deployment of ADM systems currently taking place. The fields of predictive policing and border management appear to be two of the most developed. However, the participation of state actors in these research projects, including national research centers such as KEMEA, the Hellenic Police, and various ministries, including the Hellenic Ministry of Defense, is particularly noticeable. It is important to mention that through these projects valuable technical expertise is acquired by the partners involved, which at a later stage could potentially be applied in real-life case scenarios.
At the same time, projects that focus on the use of ADM in the fields of healthcare, transportation, and medicine are in line with the National Digital Strategy of Greece and the aim is for these projects to assist the country in kick-starting its economic development. Such projects could create the infrastructure needed to boost the economy and make Greek stakeholders in these fields more competitive.