ID;Name;Website;Description;Geographical extent;Geographical coverage (country);NUTS 2021;Responsible organisation;Responsible organisation category;Functions of Government (COFOG level I);Functions of Government (COFOG level II); Status;Start Year;End Year;Process type;Application type;Cross Border;Cross Sector;Technology;Interaction;Improved Public Service;Personalized Services;Public (citizen)-centered services;Increase quality of PSI and services;More responsive, efficient, and cost-effective public services;New services or channels;Improved Administrative Efficiency;Cost-reduction;Responsiveness of government operation;Improved management of public resources;Increased quality of processes and systems;Better collaboration and better communication;Reduced or eliminated the risk of corruption and abuse of the law by public servants;Enabled greater fairness, honesty, equality;Open government capabilities;Increased transparency of public sector operations;Increased public participation in government actions and policy making;Improved public control and influence on government actions and policies;Source;AI Classification (I);AI Classification Subdomain (II) (main);AI Classification Subdomain (II) (Other I);AI Classification Subdomain (II) (Other II);AI Classification Subdomain (II) (Other III);AI Keywords;Date-updated
PSTW-1454;Classification and automatic sorting of CEMs at the Italian National Institute for Social Security (INPS);https://www.agendadigitale.eu/cittadinanza-digitale/inps-ecco-come-usiamo-lia-a-vantaggio-dei-cittadini/;One of the most important channels by which citizens can communicate with INPS is via certified email (CE). On a practical level, when the INPS operator received the CE, he reads the text and any attachments, understands the subject matter (e.g., pensions, civil invalidity, redundancy payment), and then sorts it to the official in charge of answering the subject. At the beginning of 2021, INPS started experimenting with the Open Source version of GPT-2, and then moved on to BERT, another Machine Learning model that also leverages the Transformers architecture and was developed by Google. The entire system was developed and executed in-house in the INPS Data Centres, precisely to fully satisfy GDPR and confidentiality requirements related to the sensitivity of citizens' data. It has accelerated the process of citizen communication with the Institution, it is active in 10 Italian cities of varying size and geographic area, including Rome, Milan and Naples and it has currently processed more than 1 million CEs. When fully operational, counting all territorial offices, it is estimated that between 30,000 and 40,000 working days per year will be saved. INPS was awarded by IRCAI, UNESCO's International Research Centre for Artificial Intelligence, in the world Top 10 projects supporting the 17 UN Sustainable Development Goals (SDGs).;National;Italy;IT;National Institute for Social Security (INPS);Central-government;Social protection;Social protection n.e.c.;Implemented;2021;;Internal management;Internal primary processes;No;No;Artificial intelligence;G2G;;;;;;;x;x;x;;x;x;;;;;;;https://www.agendadigitale.eu/cittadinanza-digitale/inps-ecco-come-usiamo-lia-a-vantaggio-dei-cittadini/;Learning;Machine learning;Generative AI;Optimisation;;text classification;04/10/2024
PSTW-1456;Virtual Assistant based on generative AI at INPS;https://www.inps.it/it/it/inps-comunica/atti/circolari-messaggi-e-normativa/dettaglio.circolari-e-messaggi.2023.07.messaggio-numero-2659-del-14-07-2023_14221.html;INPS has recently launched an experiment with the objective to integrate within the INPS Virtual Assistant the same 'Large Language Model' (LLM) algorithm at the basis of ChatGPT, thanks to the collaboration of Microsoft Azure with OpenAI. The aim is to offer increasingly accurate answers to citizens of certain services such as the 'Opzione Donna' pension scheme and other services.;National;Italy;IT;National Institute for Social Security (INPS);Central-government;Social protection;Social protection n.e.c.;Implemented;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;x;x;x;x;x;;;;;;;;;;;;;https://www.agendadigitale.eu/cittadinanza-digitale/inps-ecco-come-usiamo-lia-a-vantaggio-dei-cittadini/;Learning;Machine learning;Generative AI;Natural language processing;;chatbot;04/10/2024
PSTW-1735;ChatGPT as a tool for Education;https://www.agendadigitale.eu/scuola-digitale/chi-ha-paura-di-chatgpt-a-scuola-impariamo-piuttosto-a-collaborarci/;ChatGPT shows us that AI is able to provide us with information on the basis of a question, returning a coherent picture, but lacks the ability to contextualize so it does not provide us with a text that opens up perspectives that are not purely descriptive. Only man can do this, if he knows semiotic textology;National;Italy;IT;Accademia Belle Arti - Roma;Academic-Research;Education;Education not definable by level.;Planned;2022;;Internal management;Internal primary processes;No;No;Artificial Intelligence;G2C;x;;x;x;;;;;;;;;;;;;;;https://www.agendadigitale.eu/scuola-digitale/chi-ha-paura-di-chatgpt-a-scuola-impariamo-piuttosto-a-collaborarci/;Learning;Machine learning;Generative AI;;;;04/10/2024
PSTW-1745;Ministry of Justice will use ChatGPT technology to respond to citizens;https://www.publico.pt/2023/02/17/tecnologia/noticia/ministerio-justica-vai-usar-tecnologia-chatgpt-responder-cidadaos-2039270;The Ministry of Justice is working on a chatbot (chat program) to ask citizens and companies questions about legal proceedings in Portugal. The mission of the GPJ (acronym for Practical Guide to Access to Justice), presented this Friday at the Campus of Justice, is to simplify access to information with a more accessible language. The basis of the technology is the GPT-3.5 language model, the same one that is used by OpenAI's ChatGPT.;National;Portugal;PT;Ministry of Justice;Central-Government;Public order and safety;Law courts;In development;2023;;Public services and engagement;Engagement management;No;Yes;Artificial Intelligence;G2C;x;x;x;x;;;;;;;;;;;x;x;;;https://www.publico.pt/2023/02/17/tecnologia/noticia/ministerio-justica-vai-usar-tecnologia-chatgpt-responder-cidadaos-2039270;Learning;Machine learning;Generative AI;Natural language processing;;;04/10/2024
PSTW-1838;The City of Helsinki is investigating the benefits of generative artificial intelligence in various work tasks;https://www.hel.fi/en/news/the-city-of-helsinki-is-investigating-the-benefits-of-artificial-intelligence-in-various-work-tasks;This year, a group of City of Helsinki employees will experiment with the use of the new Microsoft Copilot artificial intelligence tool. Copilot is an integrated tool for applications in the M365 work environment that uses generative artificial intelligence and machine learning. Copilot can be used to manage files, create presentations and ease email overload, for example.;Local;Finland;FI;City of Helsinki ;Local Government;General public services;R&D general public services.;Pilot;2024;;Internal management;Internal support processes;No;No;Artificial Intelligence;G2G;;;;;;;x;;;;x;;;;;;;;https://www.hel.fi/en/news/the-city-of-helsinki-is-investigating-the-benefits-of-artificial-intelligence-in-various-work-tasks;Learning;Machine learning;Generative AI;;;;07/10/2024
PSTW-1850;"""LLMoin"" pilot project to provide insight into Hamburg's medium-term LLM strategy";https://hamburg-business.com/en/news/hamburgs-local-government-testing-large-language-model;"Around 100 people are taking part in the ""LLMoin"" pilot project launched by Hamburg's local government to test the use of  LLMs, which functions employees accept and the technical and organisational requirements. Large Language Models (LLMs) understand and generate human language and have been on the rise since ChatGPT at the latest. Based on artificial intelligence, LLMs can summarise texts at lightning speed and assist with research and writing texts. The AI pilot project is based on the Luminous language model generated and hosted in Germany by AlephAlpha. The AI start-up developed the front-end and data processing especially for Hamburg. LLMoin is financed by the InnoTecHH fund. The Senate Chancellery is helping authorities all over Hamburg to develop and test innovative ideas for the use of AI and other new technologies. The Ministry for IT and Digitalisation has provided funding and expertise for this since the beginning of 2023.";Local;Germany;DE;City of Hamburg ;Local Government;General public services;R&D general public services.;Pilot;2024;;Internal management;Internal support processes;No;No;Artificial Intelligence;G2G;;;;;;;x;;;;x;;;;;;;;https://hamburg-business.com/en/news/hamburgs-local-government-testing-large-language-model;Learning;Machine learning;Generative AI;;;;10/10/2024
PSTW-1899;Practical, Learning-Based Tools for Finding and Fixing Bugs;https://cordis.europa.eu/project/id/101155832;Software bugs are a major problem for software developers and users alike, as they cause crashes, security vulnerabilities, and data loss. The LearnBugs ERC project, on which this proposal is based, has developed ground-breaking techniques to automatically find bugs and to propose suitable bug fixes. These techniques are based on artificial intelligence and deep learning, making them particularly powerful for kinds of bugs missed by traditional software developer tools. This Proof of Concept proposal, named BugGPT, aims to make learning-based techniques for finding and fixing software bugs practical and usable by software developers. The project will develop practical tools that enable software developers to automatically find and fix bugs in their code. Overall, BugGPT has the potential to make a significant impact on the software development industry by making learning-based bug finding and fixing practical for software developers. If successful, the project could be the beginning of a commercial product that stirs up the market of software development tools.;National;Germany;DE;University of Stuttgart;Academic-Research;Economic affairs;R&D economic affairs;Pilot;2024;2025;Analysis, monitoring and regulatory research;Information analysis processes;No;No;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://cordis.europa.eu/project/id/101155832;Learning;Machine learning;Generative AI;Deep learning;;;06/05/2024
PSTW-1910;How Iceland is using GPT-4 to preserve its language;https://openai.com/customer-stories/government-of-iceland;Iceland partnered with OpenAI to investigate how to improve the Icelandic language abilities of its new model, GPT-4, which was officially announced today. This cooperation results from the visit of the President of Iceland and ministers to the United States in Spring 2022 when, among others, they met Sam Altman, the founder and CEO of OpenAI. One of the key goals of this collaboration was to begin building a resource that could serve to promote the preservation of other low-resource languages, helping to ensure all languages and cultures are represented in digital technologies.;National;Iceland;IS;Language Planning Department of the Government of Iceland;Central-Government;Recreation, culture and religion;Cultural services.;In development;2023;;Analysis, monitoring and regulatory research;Innovating Public Policy ;No;No;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;;;;;https://www.government.is/diplomatic-missions/embassy-article/2023/03/14/Head-start-for-Icelandic/;Learning;Machine learning;Generative AI;Natural language processing;;;10/10/2024
PSTW-1913;Generative AI Transformation Project with Derby City Council and ICS.AI;https://www.ics.ai/post/ics-ai-announces-7-million-generative-ai-transformation-project-with-derby-city-council;ICS.AI, a leader in UK public sector AI transformation, accounced its expanded partnership with Derby City Council in an innovative £7 million AI project. This 4-year initiative, launching in 2024, is a pivotal advancement in the Council's strategy, deploying the power of generative AI to transform customer service operations. Across Customer Services, the capabilities of Darcie and Ali, the two digital assistants, will be extended to become generative AI copilots, further streamlining the Council's customer service, and allowing for more complex and nuanced interactions with residents.;Local;United Kingdom;UK;Derby City Council;Local Government;General public services;General public services n.e.c.;Pilot;2024;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2C;x;x;x;x;x;x;;;;;;;;;;;;;https://govlaunch.com/projects/announcing-7m-generative-ai-transformation-project-with-derby-city-council-and-ics-ai;Learning;Machine learning;Generative AI;;;;09/10/2024
PSTW-1939;Politsei.ee large-language model LLM trained on data;https://www.kratid.ee/kasutuslood-kratid;In order to improve the findability of the necessary information on politsei.ee page, a pilot project was carried out on the data of the State Gazette to train large language models and thus create a customized model. As part of the pilot project, testing was carried out in the Azure Open AI studio. The language model on which the ChatGPT 3.5 Turbo major language model was tested. As a result of the project, it appeared that the model works worse in Estonian than in English, although the input data was in Estonian.;National;Estonia;EE;Ministry of Economic Affairs and Communications;Central-Government;Economic affairs;Economic affairs n.e.c;Pilot;2023;;Internal management;Internal primary processes;No;Yes;Artificial Intelligence;G2G;;;;;;;x;x;x;x;x;;;;;;;;https://www.kratid.ee/kasutuslood-kratid;Learning;Machine learning;Generative AI;Natural language processing;;;10/10/2024
PSTW-1944;Digital Europe Language Tools - eBriefing;https://language-tools.ec.europa.eu/EBriefingServices/Briefing;"eBriefing is an AI-based service to assist in drafting official briefings for EU Staff, but also for anyone who needs a general briefing on given topics in a similar format. Users can upload a series of documents on the subject in question and get back an AI-generated briefing in the EU Institutions’ preferred format, that is, with a basic summary, the line-to-take, defensive points, etc.  eBriefing output, like all AI-generated output, needs to be checked for accuracy and style before passing it on. eBriefing delivers a first draft to simplify the process, not a final document. Can I use it in any EU language? eBriefing is connected to eTranslation, so you can submit and receive documents in any EU language along with the others eTranslation supports. How does eBriefing know what to include in its draft?
eBriefing uses the power of the latest generation GPT models, but to get a proper briefing it works with the documents, you, the user, upload. These could be older briefings, general information on a topic, information about the upcoming event for which the briefing is needed, etc. Extensive tests have been carried out to develop a prompt that generates the output in the format usually used at the EU institutions, but with a little tweaking, the output can be adapted to other countries’ conventions or just turned into an informal summary of several documents. Especially when used for official purposes, the output needs to be reviewed and checked. What document types can it handle? eBriefing accepts docx, txt or PDF files. Only PDF files containing text will be processed. If they contain a scanned document, eBriefing cannot process them. It returns a Word document.";Across Countries;Belgium;EU;European Commission;Central-Government;General public services;General public services n.e.c.;Implemented;2023;;Internal management;Internal support processes;Yes;Yes;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;;;;;https://language-tools.ec.europa.eu/EBriefingServices/Briefing;Learning;Machine learning;Generative AI;;;;10/10/2024
PSTW-1993;Dublin City Council and OpenAI Announce Partnership to Show the Potential of AI to Support Europe’s Tourism Industry;https://irishtechnews.ie/dublin-city-council-and-openai-ai-europes-tourism/;Co-development of a Dublin Itinerary Planner Proof of Concept “A Day in Dublin”: Leveraging the advanced capabilities of GPT-4, this innovative tool will showcase how bespoke travel recommendations can make Dublin's rich cultural heritage and vibrant city life more accessible to visitors. This prototype has been designed to show the potential of generative AI to support unique, bespoke experiences for visitors instead of generic to-do lists.;Local;Ireland;IE;Dublin City Council;Local Government;Recreation, culture and religion;Cultural services.;In development;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;x;x;x;x;x;;;;;;;;;;;;;https://www.dublincity.ie/news/dublin-city-council-and-openai-announce-partnership-show-potential-ai-support-europes-tourism;Learning;Machine learning;Generative AI;;;;09/10/2024
PSTW-1996;ChatGPT-powered GOV.UK chatbot on trial;https://www.publictechnology.net/2024/01/23/uncategorized/were-not-moving-fast-and-breaking-things-chatgpt-powered-gov-uk-chatbot-trials-show-promise-but-uncover-issues-of-accuracy-and-reliability/;The Government Digital Service has pledged to take a “measured” approach to its work on a proposed GOV.UK chatbot, after public trials of the tool – powered by the technology behind ChatGPT – showed promise but shed light on “issues of accuracy and reliability”. A survey of their experiences found that about 65% of participants were satisfied overall, while 70% said they received useful answers from the technology – which, using large language model (LLM) tech from ChatGPT creator OpenAI – is designed to scout content on GOV.UK and use it provide “a human-like response” to users’ questions.;National;United Kingdom;UK;UK Government Digital Service;Central-Government;General public services;General public services n.e.c.;In development;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;x;x;x;x;x;;;;;;;;;;;;;https://insidegovuk.blog.gov.uk/2024/01/18/the-findings-of-our-first-generative-ai-experiment-gov-uk-chat/;Learning;Machine learning;Generative AI;Natural language processing;;chatbot;06/05/2024
PSTW-2001;UrbanistAI, a generative AI platform for participatory planning and co-design: Helsinki use case;https://urbanistai.com/;The City of Helsinki organised two events with UrbanistAI with citizens and entrepreneurs to study what functions they would prefer to have in the city areas in question. Generative AI was used to produce numerous alternative visions. The city officials and consultants acted as moderators for the workshop.;Local;Finland;FI;City of Helsinki;Local Government;Housing and community amenities;Community development.;Pilot;2023;;Public services and engagement;Engagement management;No;No;Artificial Intelligence;G2C;x;x;x;;x;x;;;;;;;;;x;x;x;x;https://urbanistai.com/;Learning;Machine learning;Generative AI;;;;08/05/2024
PSTW-2020;Albania to speed up EU accession using ChatGPT;https://www.euractiv.com/section/politics/news/albania-to-speed-up-eu-accession-using-chatgpt/;The aim of the project led by the Albanian government, overseen by Prime Minister Rama, is to utilize ChatGPT, an AI language model developed by Mira Murati and her team at OpenAI, to translate and integrate EU legal measures and provisions into Albanian. This initiative seeks to expedite paperwork processes and facilitate quicker accession to the EU for Albania. By leveraging AI technology, the project aims to streamline the alignment of local legislation with EU rules, eliminating the need for extensive manual translation and legal analysis, thus speeding up the process of EU integration.;National;Albania;ALB;Albanian government;Central-Government;General public services;Executive and legislative organs, financial and fiscal affairs, external affairs.;Planned;2024;;Analysis, monitoring and regulatory research;Innovating Public Policy ;No;No;Artificial Intelligence;G2G;;;;;;;x;;x;;x;;;;;;;;https://www.euractiv.com/section/politics/news/albania-to-speed-up-eu-accession-using-chatgpt/;Learning;Machine learning;Generative AI;;;;06/05/2024
PSTW-2036;CINDERELLA - Clinical Validation of an AI-based approach to improve the shared decision-making process and outcomes in Breast Cancer Patients proposed for Locoregional treatment;https://cordis.europa.eu/project/id/101057389;Treatment of breast cancer usually involves surgery that varies depending on the individual tumour, the size and shape of the breasts or personal wishes. The EU-funded CINDERELLA project proposes to create an easy-to-use application based on AI that uses a clinical database and performs picture and biometric matching to create images that simulate to the patient the outcome of surgery. The platform called CINDERELLA APProach will reinforce the role of the patient in clinical decision-making.;Across Countries;Portugal;PT;Champalimaud Foundation;Academic-Research;Health;Public health services.;Pilot;2022;2026;Public services and engagement;Service personalisation (various);Yes;No;Artificial Intelligence;G2C;x;x;x;x;;x;;;;;;;;;;;;;https://cordis.europa.eu/project/id/101057389;Learning;Machine learning;Generative AI;;;;07/05/2024
PSTW-2109;Machine Learning in Simulation for Imaging Improvement; https://www.iosb.fraunhofer.de/en/competences/optronics/visual-inspection-systems/services/machine-learning.html;"Use of the Federal Armed Forces Machine Learning in Sim. for image enhancement phase 1 As part of the ‘Machine Learning in Simulation for Imaging Improvement’ project (MLSim-BV), R & T studies for image improvement will be carried out using artificial intelligence (AI) both in visible light and infrared (for thermal simulation). A visual enhancement of the widely used Virtual Battlespace (VBS) 4 software with the aim of photorealism should be tested. Furthermore, the infrared spectral range based on RGB shall be artificially generated by image data. In the long term, the aim is to generate various AI algorithms adapted to the respective simulation data, leading to higher photorealism of the simulation. ";National;Germany;DE;Bundesamt für Ausrüstung, Informationstechnik und Nutzung der Bundeswehr;Academic-Research;Defence;Defence n.e.c.;In development;2022;;Internal management;Internal primary processes;Yes;No;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;;;;;https://ted.europa.eu/udl?uri=TED:NOTICE:634519-2022:TEXT:DE:HTML;Learning;Machine learning;Generative AI;Computer vision;;;10/10/2024
PSTW-2115;PORO an LLM for Finland;https://www.silo.ai/blog/poro-a-family-of-open-models-that-bring-european-languages-to-the-frontier;"SiloGen is today announcing a release of the first model checkpoints of a family of multilingual open source large language models (LLMs), covering all official European languages and code. Together with the University of Turku and HPLT, SiloGen launched at the end of August an initiative to build open multilingual LLMs, with the aim of ensuring European digital sovereignty and democratizing access to LLMs.
The unique open source initiative combines a world-class team, access to a record amount of compute on Europe’s most powerful supercomputer LUMI, a record amount of data, and a distinctive software layer to train LLMs.
Two months after initiating the training efforts for a family of models, we are excited to release the first checkpoint milestones for Poro 34B.
Named ‘Poro’ after the Finnish word for reindeer, this new 34 billion parameter LLM for English, Finnish and code is an early look at what is in store from our multilingual model family. Future Poro releases will expand support to other European languages and add capabilities, such as updated model architecture, expanded context window, modalities etc. ";National;Finland;FI;SILO.AI;Academic-Research;General public services;R&D general public services.;Pilot;2024;;Internal management;Internal support processes;Yes;Yes;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;;;;;https://www.silo.ai/blog/poro-a-family-of-open-models-that-bring-european-languages-to-the-frontier;Learning;Machine learning;Generative AI;Natural language processing;;;10/10/2024
PSTW-2116;Modello Italia, the first Italian Foundational Large Language Model;https://www.igenius.ai/blog/igenius-and-cineca-announce-modello-italia;iGenius and Cineca are creating a new Foundational Large Language Model to help companies and Public Administration leverage the benefits of Generative Artificial Intelligence in key areas such as healthcare, finance, and national security. Guided by the highest compliance, privacy and national security regulations. The agreement signed by iGenius and Cineca will lead to the development of a next-generation Foundational Large Language Model (“Foundational LLM”) - a neural network model that's at the core of Artificial Intelligence tools like ChatGPT or Crystal. This model will be released under an open-source license, trained with several trillion tokens and focusing on businesses and Public Administration, in Italy and Europe.;National;Italy;IT;CINECA;Academic-Research;General public services;R&D general public services.;In development;2024;;Internal management;Internal support processes;Yes;Yes;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;;;;;https://www.breakinglatest.news/technology/story-of-8-artificial-intelligences-that-speak-and-think-in-italian/;Learning;Machine learning;Generative AI;Natural language processing;;;10/10/2024
PSTW-2117;Usage of Generative Ai in Spain justice;https://rm.coe.int/20240220-javier-hernandez-cepej-ia-gen-present-and-future-v-final/1680ae9ae9;Is generative artificial Intelligence (AI) suitable for the administration of justice? The use of generative AI in the private sector is on the rise and new systems are rolled out nearly every day. But what is the situation for the public sector? What are the tools currently used and what are their limits, what are suitable areas of application? Following the “Information note on the Use of Generative AI by judicial professionals in a work-related context”, prepare by the CEPEJ Working group on Cyberjustice and Artificial Intelligence (CEPEJ-GT-CYBERJUST), the European Cyberjustice Network (ECN) presented and discussed concrete generative AI tools from Portugal and Spain applied in the public sector for the communication with citizens on legal matters and the summarising of texts.;Across Countries;Spain;ES;Ministry of Justice;Central-Government;Public order and safety;Law courts;Pilot;2023;;Internal management;Internal support processes;Yes;No;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;;;;;https://www.coe.int/en/web/cepej/-/ecn-webinar-7/2023-generative-ai-in-the-field-of-justice;Learning;Machine learning;Generative AI;Knowledge representation;Natural language processing;;10/10/2024
PSTW-2118;Vël'OK: Cycling for Survcoins;https://www.tageblatt.lu/nachrichten/luxemburg/velok-radeln-fuer-survcoins-wird-belohnt/;The virtual currency Survcoin is intended to support efforts to reduce CO2 emissions by being used to reward the users of Vël'Ok rental bikes. Nine municipalities between Differdange and Dudelange in the south of Luxembourg are affiliated with the bicycle rental service Vël'Ok. It currently has 120 stations with a total of 450 electrically assisted bicycles. Use is free of charge after prior registration with the respective municipality or CIGL. The Survcoin (short for “survival coin”) is an alternative currency based on blockchain technology. To participate, the users must first download the Survcoin app onto the smartphone. After registration, the user receives a code with which can log in to the Vël'Ok operator CIGL. When counting the kilometers, no location software is used, but the movements between the individual stations are extrapolated into kilometers.;Regional;Luxembourg;LU;Vël'Ok;Community led;Environmental protection;Pollution abatement.;Pilot;2021;;Public services and engagement;Payments and international transactions ;No;No;Blockchain;G2C;x;;x;x;;x;;;;;;;;;;;;;https://www.survcoin.lu/en/vel-ok-regional/;Learning;Machine learning;Generative AI;Automated reasoning;Natural language processing;;10/10/2024
PSTW-2119;Generative Ai for German Public Administration;https://www.linkedin.com/posts/govtechcampusde_aleph-alpha-raises-a-total-investment-of-activity-7127283659346190338-arnO?utm_source=share&utm_medium=member_desktop;The German federal government, as well as Leander and more than 10.000 city governments are harnessing the opportunities of AI to improve services for the public and public administration.;National;Germany;DE;Aleph Alpha;Academic-Research;General public services;R&D general public services.;Pilot;2023;;Internal management;internal primary processes;Yes;No;Artificial Intelligence;G2G;;;;;;;x;x;;x;x;;;;;;;;https://aleph-alpha.com/luminous-european-ai-closes-gap-to-world-leaders/;Learning;Machine learning;Generative AI;Automated reasoning;Natural language processing;;10/10/2024
PSTW-2120;Leveraging Large Language Models for Topic Classification in the Domain of Public Affair ;https://arxiv.org/abs/2306.02864;The analysis of public affairs documents is crucial for citizens as it promotes transparency, accountability, and informed decision-making. It allows citizens to understand government policies, participate in public discourse, and hold representatives accountable. This is crucial, and sometimes a matter of life or death, for companies whose operation depend on certain regulations. Large Language Models (LLMs) have the potential to greatly enhance the analysis of public affairs documents by effectively processing and understanding the complex language used in such documents. In this work, we analyse the performance of LLMs in classifying public affairs documents. As a natural multi-label task, the classification of these documents presents important challenges. In this work, we use a regex-powered tool to collect a database of public affairs documents with more than 33K samples and 22.5M tokens. Our experiments assess the performance of 4 different Spanish LLMs to classify up to 30 different topics in the data in different configurations. The results shows that LLMs can be of great use to process domain-specific documents, such as those in the domain of public affairs;National;Spain;ES;BiDA - Lab, Universidad Autónoma de Madrid (UAM);Academic-Research;General public services;R&D general public services.;Pilot;2023;;Internal management;internal primary processes;Yes;No;Artificial Intelligence;G2G;;;;;;;x;x;;x;x;;;;;;;;https://arxiv.org/pdf/2306.02864;Learning;Machine learning;Generative AI;Natural language processing;;;10/10/2024
PSTW-2125;UrbanistAI, a generative AI platform for participatory planning and co-design: Jyväskylä use case;https://urbanistai.com/;This workshop was organised by the City of Jyväskylä to improve public spaces in the city centre and reimagine the future design of the city's high street, with the help of generative AI.;Local;Finland;FI;City of Jyväskylä ;Local Government;Housing and community amenities;Community development.;Pilot;2022;;Public services and engagement;Engagement management;No;No;Artificial Intelligence;G2G;;;;;;;x;;;;x;x;;x;x;x;x;;https://urbanistai.com/;Learning;Machine learning;Generative AI;Planning and Scheduling;;;08/05/2024
PSTW-2126;UrbanistAI, a generative AI platform for participatory planning and co-design: Narva use case;https://urbanistai.com/;A three-step workshop series was organised by the City of Narva to make the city more appealing and livable for young people, involving students in this participatory planning process with the help of Generative AI.;Local;Estonia;EE;City of Narva;Local Government;Housing and community amenities;Community development.;Pilot;2021;;Public services and engagement;Engagement management;No;No;Artificial Intelligence;G2G;;;;;;;x;;;;x;x;;x;x;x;x;;https://urbanistai.com/;Learning;Machine learning;Generative AI;Planning and Scheduling;;;08/05/2024
PSTW-2152;Titormos - the citizens assist chatbot in the Municipality of Agrinio;https://www.agrinio.gr;"""Titormos"", from an ancient local hero of the city of Agrinio in Greece, is the new chatbot available on the municipality's official website. The Municipality of Agrinio is a municipality with a wide geographical span which includes both mountainous and lowland areas with different problems emerging. Prior to Titormos, citizens could only submit requests by calling a phone number during limited working hours on weekdays. With this specific solution, citizens can be informed about information concerning the city but also, most importantly, submit a problem that concern them and is of a local nature in matters of cleanliness, such as garbage collection, matters of city lighting, policing, management of stray animals, complaints, etc. After collecting the necessary information, Titormos saves it and displays the ticket – request ID and sends a relevant confirmation email. The bot is also connected to chatgpt and can answer free form questions from citizens. Finally, in the future, there will be a different version for city administrators such as the Mayor, which will be able to connect the chatbot directly to the database of requests and also to the Human Recourses (HR) database via AI, so that it can accept questions in natural language and give administrative responses such as request statistics. Also, for citizens who do not have access to or do not use computers, a voice portal for Citizens who can also submit a request and receive relevant information has been created and is active. ";Regional;Greece;EL;Municipality of Agrinio;Regional Government;General public services;General services.;Pilot;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;;x;;x;x;;;;;;;;;x;;x;x;Self-reporting tool;Communication;Natural language processing;Generative AI;Machine learning;;;29/05/2024
PSTW-2153;Large-scale high-engagement citizen participation with generative AI;https://innovationorigins.com/en/eindhoven-pioneers-with-ai-driven-city-consultation/;Enabling more citizen participation in governmental decision making could increase the quality of decisions. It is in any case likely to increase their adoption, if participation was genuinely conducted in pursuit of the best possible decision for the greatest public interest. Using ChatGPT a consortium from Eindhoven, The Netherlands, created an AI-mediated engagement app. This project was financed with $100k by OpenAI, after having been selected as one of out 10 projects from over 860 project submissions. After having conducted an initial trial during development with 450 people, it was launched as the start of a new type of city consultation in Eindhoven on 22-02-2024. Currently the EIndhoven library and the biggest youth center are supporting new events, aiming for a decision in December '24 on the city consultation becoming a monthly opportunity or people to discuss matters close to their heart.;Local;Netherlands;NL;Municipality of Eindhoven;Local Government;General public services;Executive and legislative organs, financial and fiscal affairs, external affairs.;Implemented;2024;;Public services and engagement;Engagement management;No;No;Artificial Intelligence;G2C;x;;x;x;;x;;;;;;;;;x;x;x;x;Self-reporting tool;Communication;Natural language processing;Generative AI;Machine learning;;;29/05/2024
PSTW-2161;Application of Generative AI to the Generalitat de Catalunya to facilitate citizens' understanding of legal texts;https://govern.cat/salapremsa/notes-premsa/622202/el-govern-utilitza-el-llenguatge-planer-per-facilitar-la-comprensio-dels-textos-juridics-a-la-ciutadania;"In the legal field, the Autonomous Entitat Autònoma del Diari Oficial i de Publications (EADOP) has as its mission, on the one hand, the official publication, through the DOGC, of the laws of the Generalitat of Catalonia, of the general provisions issued by the Government and the Administration of the Generalitat, and also of acts, announcements and other documents of the Generalitat and of other entities or persons, when it corresponds in accordance with the legal system. On the other hand, it is the mission of the 'EADOP to contribute to the knowledge and dissemination of current law applicable in Catalonia.  Very often, legal language is a barrier that prevents citizens from easily understanding legal texts. To eliminate this obstacle, the Government makes available to users of the Legal Portal of Catalonia and to the general public the summaries of Catalan law rules in plain language obtained from generative artificial intelligence.
It is already available in, the Legal Portal of the Generalitat of Catalonia summaries of the rules of Catalan law prepared with generative artificial intelligence. Artificial intelligence (AI) generated summaries of legal documents are provided for informational purposes only and should not be considered legal advice. These summaries are unofficial and do not replace information published in official newspapers and bulletins. They may contain expressions susceptible to linguistic improvement.";Regional;Spain;ES;Autonomous Entitat Autònoma del Diari Oficial i de Publications (EADOP) ;Regional Government;Public order and safety;Law courts;Implemented;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;x;;x;x;;;;;;;;;;x;;;x;Self-reporting tool;Learning;Machine learning;Generative AI;;;;30/07/2024
PSTW-2173;Alliance for Language Technologies (ALT-EDIC);https://gpt-nl.nl/gpt-nl/;A European Digital Infrastructure Consortium (EDIC) is a new mechanism for multi-country projects created under the Digital Decade Policy Programme 2030. EDICs allow Member States to pool funding and other resources in a flexible and efficient way, to invest in transformative digital projects. EDICs can also ensure common standards and interoperability. The mission of the ALT-EDIC (Alliance for Language Technologies EDIC) is to develop a common European infrastructure in Language Technologies, focusing particularly on Large Language Models. It seeks to improve European competitiveness, increase the availability of European language data and uphold Europe’s linguistic diversity and cultural richness. Coordinated by France, the ALT-EDIC counts sixteen Members States (Bulgaria, Croatia, Czechia, Denmark, France, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Netherlands, Poland, Slovenia and Spain) and seven observing Member States (Austria, Belgium, Estonia, Malta, Portugal, Romania and Slovakia).;Across Countries;European Union;EU;European Digital Infrastructure Consortium (EDIC);Central-Government;General public services;R&D general public services.;In development;2024;;Analysis, monitoring and regulatory research;Innovating Public Policy ;Yes;Yes;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://language-data-space.ec.europa.eu/related-initiatives/alt-edic_en;Learning;Machine learning;Generative AI;Natural language processing;;;31/07/2024
PSTW-2174;GPT-NL - A language model for the Dutch language and culture;https://chat.bggpt.ai/;The GPT-NL will be the Netherlands' open language model. The development of GPT-NL is in line with the General Data Protection Regulation (GDPR) and AI Regulation to arrive at a model that is not trained on sources that violate privacy or IP rights. The GPT-NL aims to strengthen the strategic autonomy and knowledge of the Netherlands in the field of AI, Data Science, and Data Spaces. Financing for the model comes from RVO/Ministry of Economic Affairs and Climate Policy. The project plan “Facility for a sovereign Dutch language model” was submitted for this in May 2023 and awarded at the end of October 2023. An amount of 13.5 million euros has been made available for the project.;National;Netherlands;NL;Ministry of Economic Affairs and Climate;Central-Government;General public services;R&D general public services.;Implemented;2023;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2B;x;;x;;;x;;;;;;;;;;;;;https://gpt-nl.nl/gpt-nl/;Learning;Machine learning;Generative AI;Natural language processing;;natural language generation;31/07/2024
PSTW-2175;BgGPT - A bulgarian LLM;https://eufarmbook.eu/it/about, https://www.tno.nl/en/digital/digital-innovations/data-sharing/generative-ai/;INSAIT – Institute for Computer Science, Artificial Intelligence and Technology is developing BgGPT, a series of state-of-the-art generative AI for the Bulgarian language, created for the Bulgarian users, institutions, public, and private organisations. BgGPT is released as a series of free and open language models.;National;Bulgaria;BG;INSAIT – Institute for Computer Science, Artificial Intelligence and Technology;Academic-Research;General public services;General public services n.e.c.;In development;2024;2029;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2C;x;x;x;;;x;;;;;;;;;;;;;https://bggpt.ai/blog/;Learning;Machine learning;Generative AI;Natural language processing;;chatbot;07/10/2024
PSTW-2176;GenAI chatbot on the EU-FarmBook platform;https://digital-strategy.ec.europa.eu/en/policies/edic;EU-FarmBook (supporting knowledge exchange between all AKIS actors in the European Union) is a Horizon Europe project that is working at regional, national, and European (EU) levels to build an Online Platform. Its main purpose is to support knowledge exchange, by further developing, expanding and maintaining an easily accessible and user-friendly EU-wide Online Platform for practitioners in the agriculture and forestry sectors. With the help of TNO, a Dutch research organisation, Generative AI is being deployed within the project's platform to construct a conversional chatbot that allows farmers, horticulturtists and policymakers to interactively retrieve relevant information.;Across Countries;European Union;EU;EU-FarmBook;Consortium;Economic affairs;Agriculture, forestry, fishing and hunting;Implemented;2022;;Public services and engagement;Service personalisation (various);Yes;No;Artificial Intelligence;G2B;x;x;x;;x;x;;;;;;;;;;;;;https://eufarmbook.eu/it/about;Learning;Machine learning;Generative AI;Natural language processing;;chatbot;31/07/2024
PSTW-2177;CitiVERSE - LDT CitiVERSE EDIC;https://www.ai.se/en/project/gpt-sw3;The Networked Local Digital Twins towards the CitiVERSE (LDT CitiVERSE EDIC) initiative aims to connect existing local digital twins across Europe, forming the basis for the EU CitiVERSE. Digital twins use data, analytics, and AI to create real-time simulation models that mirror the cities they represent. The CitiVERSE focuses on advancing generative AI applications in smart cities, including simulations addressing (among other things) the impact of changing traffic conditions on air quality, decarbonisation, and congestion. Additionally, it explores generative AI-based virtual reality applications to improve citizen interaction (for instance, simplifying the consultation and understanding of planned urban changes).  Croatia, Estonia, France, Latvia, Portugal, Slovenia and Spain firstly called on the Commission to set up the European Digital Infrastructure Consortium for Networked Local Digital Twins towards the CitiVERSE (LDT CitiVERSE EDIC). ;Across Countries;European Union;EU;LDT CitiVERSE EDIC;Consortium;General public services;R&D general public services.;Implemented;2024;;Analysis, monitoring and regulatory research;Innovating Public Policy ;Yes;Yes;Artificial Intelligence;G2C;x;;x;x;;x;;;;;;;;;x;;x;;https://digital-strategy.ec.europa.eu/en/policies/edic;Learning;Machine learning;Generative AI;;;;31/07/2024
PSTW-2178;GPT-SW3 - The first large-scale generative language model for the Swedish language;"https://www.ai.se/en/project/shared-digital-assistant-public-sector
";"AI Sweden, together with RISE and WASP WARA Media & Language, have developed a large-scale generative language model for the Nordic languages, primarily Swedish. GPT-SW3 is the first truly large-scale generative language model for the Swedish language. Based on the same technical principles as the much-discussed GPT-4, GPT-SW3 will help Swedish organisations build language applications never before possible. AI Sweden is the national center for applied AI and brings together more than 120 partners across the public and private sectors as well as academia. AI Sweden is funded by the Swedish government and our partners, both public and private. 
";National;Sweden;SE;AI Sweden;Academic-Research;Economic affairs;R&D economic affairs;In development;2024;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2B;x;x;x;;;x;;;;;;;;;x;x;;;https://www.ai.se/en/project/gpt-sw3;Learning;Machine learning;Generative AI;Natural language processing;;;31/07/2024
PSTW-2179;Svea - A shared digital assistant for the public sector;https://www.lighton.ai/fr/success-stories;The project, a shared digital assistant for the public sector, is a collaboration between Swedish authorities, municipalities, regions, and the business sector, coordinated by AI Sweden. The objective is to promote national cooperation on AI for text tasks and create conditions for shared solutions in the public sector.The ultimate goal is to support public sector employees and free up their time so they can focus on more creative and interpersonal tasks. Success requires public sector organisations to share critical resources while sharing the burden of gathering the data needed to create truly competent and flexible solutions. The first phase of this initiative focuses on data generation, change management, identifying use cases, and developing a prototype for an AI assistant using LLMs. Svea is the prototype for a shared digital assistant, which is under development in this project. The AI-based prototype will be designed to help public sector workers with text-based tasks. Currently, based on the Mixtral open language model from the French company Mistral, more models will be tested in phase 2. Employees will interact with Svea via a web-based chat. Svea can already answer business-specific questions, summarise documents, extract information, rewrite texts in easy-to-read Swedish, and much more. As of today, the following public administrations are involved in the project: Kungsbacka municipality, Tjörn municipality, Gothenburg city, Region Skåne, Region Halland, and Västra Götaland Region.;Local;Sweden;SE;Kungsbacka municipality, Tjörn municipality, Gothenburg city, Region Skåne, Region Halland, and Västra Götaland Region;Local Government;General public services;General public services n.e.c.;Implemented;2024;;Internal management;Internal support processes;No;Yes;Artificial Intelligence;G2G;;;;;;;x;;;x;x;x;;;x;x;;;https://www.ai.se/en/project/shared-digital-assistant-public-sector;Learning;Machine learning;Generative AI;Natural language processing;;;31/07/2024
PSTW-2180;LightOn's Paradigm platform - Generative AI at the Île-de-France Regional Council;https://ai4science-amsterdam.github.io/ai4smm_projects/#machine-learning-based-models-of-plant-protein-mixtures-for-sustainable-food-design;Agents in the Ile-de-France Region struggle to find the relevant information to properly use their internal software, which impacts their ability to carry out their operations, and takes up a lot of the IT team's time. The Region's Digital Transformation Hub team added the necessary technical documentation to LightOn's Paradigm platform. This allows their agents to access instant support by easily searching through their documentation. Plans are underway to extend these benefits to the general public, demonstrating the impact of AI on public service management. Paradigm, an intelligent assistant serving agents in the digital transformation hub (who are responsible for managing the information systems, infrastructures and applications for all agents at headquarters). Designed by the startup LightOn, the tool quickly provides a precise response to questions asked in natural language. To do this, it uses existing technical documentation.;Regional;France;FR;Île-de-France Regional Council;Regional Government;General public services;General services.;In development;2024;;Internal management;Internal support processes;No;No;Artificial Intelligence;G2G;;;;;;;x;;;x;x;;;;x;x;;;https://www.smartcitymag.fr/article/1378/ia-point-d-etape-sur-les-8-cas-d-usage-testes-en-ile-de-france;Learning;Machine learning;Generative AI;Natural language processing;;;10/10/2024
PSTW-2184;SOLARIS - Democratic responses to the challenges of deepfakes and infodemics;https://projects.illc.uva.nl/solaris/Project-Summary/;Generative adversarial networks (GANs) are generative AI models capable of reproducing realistic images, videos and voice outputs. Because they have the potential to create highly realistic content, they can also pose a threat to democracies through the creation and circulation of deepfakes, which spread false information and undermine trust. In this context, the EU-funded SOLARIS project will focus on the challenges and opportunities posed by GANs in the context of democracy. It will study how deepfakes circulate and are trusted by users, thus threatening democratic processes. It will also develop regulatory guidelines about GANs and generative AI for the promotion of democratic engagement. The project will involve three use cases to simulate the circulation of threatening content, and co-create value-based GANs content to enhance digital citizenship.;Across Countries;European Union;EU;University of Amsterdam;Consortium;Public order and safety;R&D public order and safety.;In development;2023;2026;Analysis, monitoring and regulatory research;Innovating Public Policy ;Yes;Yes;Artificial Intelligence;G2G;;;;;;;x;;;;;;;x;;;;;https://projects.illc.uva.nl/solaris/Project-Summary/;Learning;Machine learning;Generative AI;;;generative adversarial network;31/07/2024
PSTW-2193;TrustLLM - Democratise trustworthy and efficient LLM technology for Europe;https://cordis.europa.eu/project/id/101135724, https://luminous-horizon.eu/;The TrustLLM project will develop European large language models (LLMs) on an unprecedented scale, trained on the largest amount of text so far in European AI, covering a range of underrepresented languages, and pushing the limits of European exascale computing. The main objective is the development of an open, trustworthy, and sustainable LLM initially targeting the Germanic languages. This will create the foundation for an advanced open ecosystem for next generation modular and extensible European trustworthy, sustainable, and democratised large language models. The TrustLLM project and the surrounding ecosystem will enable, support, and improve context-aware human-machine interaction in a wide range of applications.;Across Countries;European Union;EU;Linkopings Universitet;Consortium;General public services;R&D general public services.;In development;2023;;Public services and engagement;Service personalisation (various);Yes;Yes;Artificial Intelligence;G2C;x;;x;;;x;;;;;;;;;;;;;https://cordis.europa.eu/project/id/101135671;Learning;Machine learning;Generative AI;Natural language processing;;case-based reasoning;31/08/2024
PSTW-2194;GENAI4LEX-B;https://trustllm.eu/;"The ""GENAI4LEX-B"" project was the winner for the ""Supporting the preparation of initiative acts"" area. The three proposed prototypes provide tools to learn more about the regulatory sources; simplify and summarise amendments; measure the adherence of the bills to the policies they intend to pursue. The Institute of Legal Informatics and Judicial Systems (CNR-Igsg)  participated in ‘GENAI4LEX-B’ with the support of the start-up Aptus.AI. The project was presented by an interdisciplinary consortium of universities and research centres in collaboration with a number of spin-offs, coordinated by Monica Palmirani of Alma Mater Studiorum - University of Bologna ALMA-AI. The consortium was also composed of: Luiss Guido Carli University - Centre for Parliamentary Studies in partnership with Asimov AI, University of Turin and University of Verona.";National;Italy;IT;Alma Mater Studiorum – University of Bologna ALMA-AI;Consortium;General public services;R&D general public services.;In development;2024;2026;Internal management;Internal support processes;No;No;Artificial Intelligence;G2G;;;;;;;x;;;x;x;x;;;;;;;https://www.cnr.it/it/news/12868/il-progetto-genai4lex-b-premiato-alla-camera-dei-deputati-intelligenza-artificiale-generativa-per-i-lavori-parlamentari;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2195;F13, the LLM used by the State of Baden-Württemberg in collaboration with Aleph Alpha;https://www.cnr.it/it/news/12868/il-progetto-genai4lex-b-premiato-alla-camera-dei-deputati-intelligenza-artificiale-generativa-per-i-lavori-parlamentari;"The system, F13, was developed primarily by the Heidelberg-based AI start-up Aleph Alpha, which has been touted as a German response to OpenAI's ChatGPT, though aimed at government, administration and industry in particular. The Innovation Lab Baden-Württemberg (InnoLab_bw) presents the new text assistance ""F13"" for the administration in the state. Based on generative artificial intelligence, this programme was developed to sustainably relieve the employees of the state administration. New functions are introduced to optimize workflows. The summary function allows staff to select different compression levels for pasted texts. The Cabinet Template (KV) Note converts uploaded cabinet templates into a summarised note template. The research assistant supports the preparation of large amounts of information, while the ""Vermerkomat"" enables the creation of a comprehensive flow text. The ""Vermerkomat"" is currently in the test phase. It is important to note that the summarised texts must always be checked and controlled by one person. Complete automation of the processes is not planned. Close involvement of the users plays a decisive role in the further development of the functions of ""F13"" in terms of quality. For this reason, a diverse group of testers from different areas was integrated into the feedback process at an early stage. From May 2023, all employees who have access to the state administration network will be able to use ""F13"".
";Regional;Germany;DE;State of Baden-Württemberg;Regional Government;General public services;General services.;Pilot;2023;;Internal management;Internal support processes;No;Yes;Artificial Intelligence;G2G;;;;;;;x;;;;x;x;;;;;;;https://www.yahoo.com/news/baden-w-rttemberg-first-german-170630420.html?guccounter=1;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2197;ALEIA digital assistant - Generative AI at the Île-de-France Regional Council;https://aleph-alpha.com/de/mit-dem-neuen-f13-in-die-verwaltung-der-zukunft-baden-wurttemberg-geht-mit-f13-produktiv/;The tool helps people to find regional aid, as the region's 350 or so aid schemes can leave them bewildered. The system, designed by start-up Aleia, is a combination of a search engine and a conversational agent. It uses the Mistral tricolour language model to formulate answers based on the region's corpus of documents. The data is hosted in France and the answers are continuously improved. And if it is asked for help that is not at all within the remit of the region, it gives a negative response’, points out Antoine Courert, one of the founders of Aleia.;Regional;France;FR;Île-de-France Regional Counci;Regional Government;General public services;General services.;Implemented;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;x;x;x;x;x;;;;;;;;;;;;;https://www.banquedesterritoires.fr/la-region-ile-de-france-presente-son-demonstrateur-dia-de-service-public;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2199;Open Spanish LLM model in collaboration with IBM;https://www.banquedesterritoires.fr/la-region-ile-de-france-presente-son-demonstrateur-dia-de-service-public;"In a new effort announced today, the government of Spain and IBM said they would come together to develop and deploy cutting-edge AI technologies. This will include a new open-source large language model (LLM) in Spanish. Together, they plan to revolutionise how Spanish-speaking countries and organisations can do business in the age of AI.
";National;Spain;ES;Government of Spain;Central-Government;General public services;R&D general public services.;Implemented;2024;;Analysis, monitoring and regulatory research;Innovating Public Policy ;No;Yes;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://research.ibm.com/blog/new-spanish-llm-ai;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2200;"OpenGPT-X project to create a LLM ""Made in Germany""";https://research.ibm.com/blog/new-spanish-llm-ai;With the unique expertise of project partners representing the entire generative AI value chain, OpenGPT-X aims to advance cutting-edge research while maintaining a strong commitment to business applications to drive the adoption of generative AI in the German economy. The collaborative effort across industry, research and media is funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWK) from January 2022 to December 2024. The project is funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWK) as part of the funding competition “Innovative and practical applications and data spaces in the Gaia-X digital ecosystem” from January 2022 to March 2025 (more information). The total funding amounts to about 14 million Euros.;National;Germany;DE;OpenGPT-X ;Consortium;General public services;R&D general public services.;Planned;2022;;Analysis, monitoring and regulatory research;Innovating Public Policy ;No;Yes;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://opengpt-x.de/en/about/;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2201;"Development of GenAI-based tool for education by gov.uk - ""Building a proof of concept for Generative AI feedback and resource generation in education contexts""";https://opengpt-x.de/en/;The National Institute of Teaching (NIoT) and the Department for Education (DfE) have been exploring potential applications for Generative AI in the education sector as part of a wider effort to ‘transform a teacher’s day to day work’ – reducing workload and improving educational outcomes by automating routine tasks. As part of this exploratory work, we have built a proof of concept (PoC) tool to explore the potential for using Large Language Models (LLMs) for these purposes. The tool is targeted towards supporting teachers with evaluating work against the National Curriculum and providing formative feedback and activities to pupils.;National;United Kingdom;UK;Department for Education - UK Government;Central-Government;Education;R&D education.;In development;2023;;Internal management;Internal support processes;No;No;Artificial Intelligence;G2G;;;;;;;x;x;;;x;;;;;;;;https://assets.publishing.service.gov.uk/media/66cdb0cc239c5e6b4dc0533a/Use_cases_for_generative_AI_in_education_technical_report.pdf;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2202;Developing a Generative AI platform for construction cost management;https://assets.publishing.service.gov.uk/media/66cdb0cc239c5e6b4dc0533a/Use_cases_for_generative_AI_in_education_technical_report.pdf;The Futures of Systems, People and Projects (FUSP) laboratory has been awarded a £1.18m Innovate UK grant to produce a generative AI platform for boosting the productivity and accuracy of cost planning in construction. This new research project follows on from a successful feasibility study funded by Innovate UK. We have now been awarded follow-on funding to produce a “BIM-enabled Generative AI Platform for Productivity and Accuracy Enhancement of Construction Cost Planning (BIM-GAIcost)”. The potentials of generative AI will become a useful tool to reduce cost overruns which are prevalent in 69% of projects and the 15% increase in UK construction costs. Our successful pilot project produced costplan.ai – we will now upgrade this with new features that merge a data-driven Building Information Modelling (BIM) for early cost advice through generative AI. The prompting feature of the AI digital platform will enhance productivity by up to 85% by reducing the time taken to produce initial cost plans and produce accuracy measures for UK contractors and clients. The research will enhance the web application – costplan.ai - to produce construction cost estimation documents through large language models (LLM) of cost data and support the automation of BIM cost related documents with accuracy indications to predict and avoid cost overruns in the planning phase.;National;United Kingdom;UK;Leeds Beckett University;Academic-Research;Economic affairs;Mining, manufacturing and construction.;Pilot;2024;;Analysis, monitoring and regulatory research;Prediction and planning ;No;No;Artificial Intelligence;G2B;x;;;;x;x;;;;;;;;;;;;;https://www.ukri.org/wp-content/uploads/2024/08/IUK-070824-TMF-announcement-7.8.24-project-list.pdf;Learning;Machine learning;Generative AI;Planning and Scheduling;;;31/08/2024
PSTW-2204;Amsterdam: Generating sustainable materials with Generative AI;https://www.leedsbeckett.ac.uk/blogs/school-of-beec/2024/09/ai-for-construction-cost-management/;Natural language processing – the backbone of gen AI chatbots like ChatGPT – uses word data to make new linguistic connections. Now a research project at the University of Amsterdam is swapping words for molecules to create new, sustainable materials. The AI4SMM team is initially focused on creating energy storage salts, sustainable steel, safe plastic and new plant proteins.;National;Netherlands;NL;University of Amsterdam;Academic-Research;Environmental protection;R&D environmental protection.;Pilot;2024;;Analysis, monitoring and regulatory research;Innovating Public Policy ;No;No;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://www.uva.nl/en/shared-content/faculteiten/en/faculteit-der-natuurwetenschappen-wiskunde-en-informatica/news/2024/02/shaping-the-future-of-materials-using-artificial-intelligence.html?cb;Learning;Machine learning;Generative AI;;;;31/08/2024
PSTW-2205;Qra - The first  Polish open-source LLM;https://tech.eu/2024/08/07/monumos-innovations-in-electric-drive-design-a-ps750000-boost-from-innovate-uk/;Gdańsk University of Technology and AI Lab from the Information Processing Center (OPI) - National Research Institute have developed Polish-language generative neural language models based on a terabyte of text data exclusively in Polish. Qra is the first of its kind and the best in Polish language modeling equivalent of open tools such as Meta or Mistral AI. Qra understands content in Polish better, understands questions asked in Polish better and creates coherent texts better. it is available as open-source on the huggingface platform and it is still being finetuned.;National;Poland;PL;Gdańsk University of Technology;Academic-Research;General public services;R&D general public services.;Pilot;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;;;;;x;;;;;;;;;;;;;https://huggingface.co/OPI-PG/Qra-1b;Learning;Machine learning;Generative AI;Natural language processing;;;31/08/2024
PSTW-2206;OpenLLM-Ro - The first LLM in Romanian language;https://www.weforum.org/agenda/2024/07/generative-ai-smart-cities/, https://www.uva.nl/en/shared-content/faculteiten/en/faculteit-der-natuurwetenschappen-wiskunde-en-informatica/news/2024/02/shaping-the-future-of-materials-using-artificial-intelligence.html?cb;Starting from the second part of 2023, a team of researchers from POLITEHNICA Bucharest, the University of Bucharest and the Institute of Logic and Data Science worked on the development and training of this LLM. The academic partners contributed researchers who worked pro-bono and, in addition, POLITEHNICA Bucharest also provided the computing power needed to train the model. It is available online and it is currently being finetuned.;National;Romania;RO;POLITEHNICA Bucharest, the University of Bucharest and the Institute of Logic and Data Science;Academic-Research;General public services;R&D general public services.;Pilot;2024;;Public services and engagement;Service personalisation (various);No;No;Artificial Intelligence;G2C;x;;;;;x;;;;;;;;;;;;;https://therecursive.com/romania-launches-its-first-open-source-llm-model/;Learning;Machine learning;Generative AI;Natural language processing;;;31/08/2024
PSTW-2208;GenAI-based chatbot to help the public access services on official websites;https://opi.org.pl/sukces-wspolpracy-pg-i-opi-w-obszarze-ai/;The project, which began in January, aims to deliver a “chat agent” that would supply information to the public on business supports that are being provided by the Government. Initial funding for the pilot project has been provided by international tech giant Microsoft, and it is set to be delivered in collaboration with the Office of the Government Chief Information Officer. According to the briefing note, the chatbot will work to trial the use of Generative AI to help provide information to the general public.;National;Ireland;IE;Government of Ireland;Central-Government;General public services;General services.;In development;2024;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2C;x;x;x;;;x;;;;;;;;;;;;;https://www.irishexaminer.com/news/arid-41395996.html;Learning;Machine learning;Generative AI;Natural language processing;;chatbot;31/08/2024
PSTW-2209;FinGPT-3 - The open Finnish LLM;https://openllm.ro/;The TurkuNLP Group is a group of researchers at the University of Turku as well as the UTU graduate school (UTUGS). The main focus of our research are various aspects of natural language processing / language technology and digital linguistics, ranging from corpus annotation and analysis to machine learning theory and applications. In 2023, they have trained  generative GPT-3-like models for Finnish. The models were created in collaboration with the National Library of Finland. This project has received funding from the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350.;National;Finland;FI;TurkuGPT -  University of Turku;Academic-Research;General public services;R&D general public services.;In development;2023;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2C;x;x;x;;;x;;;;;;;;;;;;;https://turkunlp.org/;Learning;Machine learning;Generative AI;Natural language processing;;;31/08/2024
PSTW-2210;"NorGPT - The open LLM ""Made in Norway""";https://www.ntnu.edu/norwai/norgpt-language-models;LLM using Norwegian Language developed by NorwAI and the Norwegian University of Science and Technology (NTNU);National;Norway;NO;NorwAI and NTNU;Academic-Research;General public services;R&D general public services.;In development;2024;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2C;x;x;x;;;x;;;;;;;;;;;;;https://www.ntnu.edu/norwai/norgpt-language-models;Learning;Machine learning;Generative AI;Natural language processing;;;31/08/2024
PSTW-2211;‘Democratic Engagement through a Public Chatbot’ (DepuChat) - How GenAI can increase transparency;https://turkunlp.org/gpt3-finnish;The Chamber of Deputies is more transparent and closer to the citizens thanks to the project ‘Democratic Engagement through a Public Chatbot’ (DepuChat), developed by the European Interdepartmental Research Centre for Advanced Studies on Digital Innovation IDEAS (Research Center IDEAS - Innovation Digital European Advanced Studies) of the University of Roma Tre in collaboration with the Department of Legal Sciences of the University of Florence. This is a proposal for a chatbot that, through generative AI, will make data and information on the Chamber, its bodies and procedures, individual MPs, and the political positions expressed by MPs and the Government more usable, including graphically. The purpose of this technological tool is to bring citizens closer to the Institutions. ;National;Italy;IT;University of Roma Tre;Academic-Research;General public services;R&D general public services.;Implemented;2024;;Public services and engagement;Service personalisation (various);No;Yes;Artificial Intelligence;G2C;x;x;x;;;x;;;;;;;;;x;x;;;https://www.wired.it/article/intelligenza-artificiale-generativa-camera-dei-deputati-anna-ascani/;Learning;Machine learning;Generative AI;Natural language processing;;chatbot;11/09/2024
PSTW-2213;Supporting Civil Servant with Generative AI;https://www.ntnu.edu/norwai/norgpt-language-models, https://www.uio.no/dscience/english/news-and-events/news/presentation-from-nordic-perspectives-on-ai/4jon-atle-gulla_norgpt_131023-np.pdf, https://www.ntnu.edu/norllm/four-models-built-four-new-ones-in-the-pipeline;"The civil service, the largest employer in France, cannot remain on the sidelines of this technological revolution, at the risk of suffering it.
In order to measure the concrete consequences, 1000 volunteer agents will be equipped with a tool using artificial intelligence to help them write online responses to users' opinions and comments, during their online procedures. Agents will supervise and modify the response before sending it to the user. The data is completely anonymised, so confidentiality is completely preserved for the user. These tools must be put at the service of agents to enable them to be more efficient on certain tasks in order to save time and thus strengthen human contact between public service agents and the French people. This experiment, led by the Interministerial Directorate for Public Transformation (DITP) and the Interministerial Digital Directorate (DINUM), will be extended, by the end of the year, to the agents of the France services houses in their support for users at the counter.  The solutions experimented is made by https://www.allobrain.com/";National;France;FR;"Ministry of Processing and Function Public
Interministerial Directorate for Public Transformation (DITP) and the Interministerial Digital Directorate (DINUM),";Central-Government;General public services;R&D public order and safety.;Implemented;2023;;Internal management;Internal primary processes;No;Yes;Artificial Intelligence;G2G;;;;;;;x;;x;;x;x;;;;;;;Stanislas GUERINI experiments with generative artificial intelligence in public services | Interdepartmental Directorate for Public Transformation (modernisation.gouv.fr);Learning;Machine learning;Generative AI;;;;11/09/2024
PSTW-2215;Legislab - monitoring legislative quality with GenAI;https://www.quotidiano.net/politica/lintelligenza-generativa-camera-deputati-44462545;"The project is designed to help and monitor, through an artificial intelligence agent, the quality of the legislative system. This analysis occurs objectively, in the sense that the laws that are to be analyzed are compared with all those issued from the Constitution onwards. The project is based on the work of building a database of Italian laws, which is constantly updated. A second part of the platform aims to help the parliamentarian or the legislator to write the laws. Through a screen it is possible to insert an input text, for example the title of a new law, and the tool searches for the articles and laws, in the entire database, that may be important to cite or look at if you want to write a law on a certain topic. It does not help to write the law, that remains the task of the expert. To avoid ""hallucinations"" the model is forced to rely only on data from other laws, objective data and linguistic metrics that derive from scientific literature.";National;Italy;IT;The Polytechnic University of Milan;Academic-Research;General public services;Executive and legislative organs, financial and fiscal affairs, external affairs.;Planned;2024;;Public services and engagement;Internal primary processes;No;Yes;Artificial Intelligence;G2G;;;;;;;x;;;;x;;;;;;;;https://www.quotidiano.net/politica/lintelligenza-generativa-camera-deputati-44462545;Learning;Machine learning;Generative AI;;;;13/09/2024
PSTW-2216;Leveraging Large Language Models for Topic Classification in the Domain of Public Affairs;https://arxiv.org/abs/2306.02864;"A study focused on leveraging Large Language Models (LLMs) for the classification of topics in public affairs documents, which is considered crucial for promoting transparency, accountability, and informed decision-making. The research involves collecting over 33,000  Spanish text corpora of public affairs documents, using a regex-powered tool to process and annotate legislative initiatives from the Spanish Parlament during a capture period over 2 years. The documents are annotated with multiple labels, reflecting a variety of topics defined by experts in public affairs regulation. The main goal is to assess the performance of four different Spanish LLMs in classifying up to 30 different topics from these documents. 
The experiments show that LLMs are potent tools for processing domain-specific documents, such as those in public affairs, with the combination of a Large Language Model backbone and SVM classifiers yielding accuracies over 85%.
The work has also described a system architecture used for automatic document analysis in public affairs, which includes a Harvester module for data collection, a Document Layout Analysis module, and a Text Processing module that utilizes LLM technology for multi-label topic classification.
Overall, the study contributes a novel functionality for multi-label topic classification within a general document analysis system, presents a new dataset of annotated public affairs documents, and provides experimental evidence of the effectiveness of LLMs in this domain.";National;Spain;ES;BiDA - Lab, Universidad Autónoma de Madrid;Academic-Research;General public services;General services.;Pilot;2024;;Public services and engagement;Internal support processes;Yes;No;Artificial Intelligence;G2G;;;;;;;x;;;x;;x;;;x;x;x;;https://arxiv.org/abs/2306.02864;Learning;Machine learning;Generative AI;;;;26/09/2024
PSTW-2218;ITALIAN-LEGAL-BERT: A Pre-trained Transformer Language Model for Italian Law;https://ceur-ws.org/Vol-3256/km4law3.pdf;The state of the art in natural language processing is based on transformer models that are pre-trained on general knowledge and enable efficient transfer learning in a wide variety of downstream tasks even with limited data sets. However, these models significantly decrease performance when operating in specific and sectoral domains. This is problematic in the Italian legal context, as there are many discrepancies between the language found in generic open source corpora (e.g., Wikipedia and news articles) and legal language, which can be cryptic, Latin-based, and domain idiolectal formulas. In this paper, we introduce the ITALIAN-LEGAL-BERT model with additional pre-training of the Italian BERT model on Italian civil law corpora. It achieves better results than the ‘general-purpose’ Italian BERT in different domain-specific tasks;Across Countries;Italy;IT;Sant’Anna School of Advanced Studies;Academic-Research;General public services;Executive and legislative organs, financial and fiscal affairs, external affairs.;Pilot;2022;;Internal management;Engagement management;Yes;Yes;Artificial Intelligence;G2C;x;;;;;x;;;;;;;;;;;;;https://ceur-ws.org/Vol-3256/km4law3.pdf;Learning;Machine learning;Generative AI;;;;26/09/2024
PSTW-2219;LEGAL-BERT: The Muppets straight out of Law School;https://aclanthology.org/2020.findings-emnlp.261/;BERT has achieved impressive performance in several NLP tasks. However, there has been limited investigation on its adaptation guidelines in specialised domains. Here we focus on the legal domain, where we explore several approaches for applying BERT models to downstream legal tasks, evaluating on multiple datasets. Our findings indicate that the previous guidelines for pre-training and fine-tuning, often blindly followed, do not always generalize well in the legal domain. Thus we propose a systematic investigation of the available strategies when applying BERT in specialised domains. These are: (a) use the original BERT out of the box, (b) adapt BERT by additional pre-training on domain-specific corpora, and (c) pre-train BERT from scratch on domain-specific corpora. We also propose a broader hyper-parameter search space when fine-tuning for downstream tasks and we release LEGAL-BERT, a family of BERT models intended to assist legal NLP research, computational law, and legal technology applications.;Across Countries;Greece;EL;Athens university of economics and business;Academic-Research;General public services;Executive and legislative organs, financial and fiscal affairs, external affairs.;Pilot;2020;;Internal management;Engagement management;Yes;Yes;Artificial Intelligence;G2C;x;;;;;x;;;;;;;;;;;;;https://aclanthology.org/2020.findings-emnlp.261/;Learning;Machine learning;Generative AI;;;;26/09/2024
PSTW-2221;Predictice - Summaries of Court Decisions;https://bit.ly/3Y6whXk;This French legal research product uses ChatGPT to generate automatic summaries of court decisions. GPT-3 (Generative Pretrained Transformer 3) is a state-of-the-art language model that can generate text autonomously using machine learning algorithms. This tool is considered very promising. Predictice provides access to 26 million documents, such as court decisions. These are texts that are several pages long - and often even dozens of pages. They are therefore complicated to read and sometimes turn out to be uninteresting after several minutes of reading. The Predictice teams realized that while ChatGPT frequently failed when trying to craft legal responses, this was almost never the case when it came to synthesizing documents.;Across Countries;France;FR;https://predictice.com/;Private sector;Public order and safety;R&D public order and safety.;Implemented;2023;;Analysis, monitoring and regulatory research;Information analysis processes;No;No;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://bit.ly/3Y6whXk;Learning;Machine learning;Generative AI;;;;26/09/2024
PSTW-2222;Henchman - contract drafing with AI support;https://bit.ly/3Zx0r7I;Uses GPT-3.5 to enrich drafting options, for example to change single to plural in a clause or add an element to a clause.;Across Countries;Belgium;BE;https://henchman.io/;Private sector;Public order and safety;R&D public order and safety.;Implemented;2023;;Analysis, monitoring and regulatory research;Information analysis processes;No;No;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://bit.ly/3Zx0r7I;Learning;Machine learning;Generative AI;;;;26/09/2024
PSTW-2223;Robin AI - LLM for supporting Legals;https://cutt.ly/gwijcRyQ;"RobinAI is a legal industry platform that uses Large Language Models (LLMs), including Anthropic's model ""Claude  to automate tasks such as reading, writing, and editing text data. The platform integrates these models to analyse contracts and suggest real-time edits based on clients' preferences	 enabling efficient review and approval processes. While acknowledging the potential for AI errors""";Across Countries;United Kingdom;UK;https://www.robinai.com/;Private sector;Public order and safety;R&D public order and safety.;Implemented;2023;;Analysis, monitoring and regulatory research;Information analysis processes;No;No;Artificial Intelligence;G2B;x;;;;;x;;;;;;;;;;;;;https://cutt.ly/gwijcRyQ;Learning;Machine learning;Generative AI;;;;26/09/2024
PSTW-2224;Harvey - LLM for supporting contract analysis;https://bit.ly/3k6w2xv;Harvey, the tool rolled out by Allen & Overy, apparently uses GPT-4 to help lawyers automate contract analysis, due diligence, conduct research, and generate insights, recommendations, and predictions across multiple practice areas.;Across Countries;United Kingdom;UK;https://www.harvey.ai/;Private sector;Public order and safety;R&D public order and safety.;Implemented;2023;;Analysis, monitoring and regulatory research;Information analysis processes;No;No;Artificial Intelligence;G2B;x;x;;;;x;;;;;;;;;;;;;https://bit.ly/3k6w2xv;Learning;Machine learning;Generative AI;;;;26/09/2024