European news

Opportunities and risks of artificial intelligence Chat GPT technology

Recently ChatGPT has gained prominence. It is a free to use AI system. Chat GPT is an AI chatbot that uses GPT’s language model to interact with humans in a conversational way. It’s optimized for dialogue by human trainers and the free version relies on 20 billion of GPT-3’s parameters to produce convincing textual responses.  GPT-3 and GPT-4, for example, were trained on an unbelievable quantity of written work in order to produce texts. It basically amounts to the whole of the public internet, plus hundreds of thousands of books, articles, and other documents.  BgGPT is tailored to the specifics of the Bulgarian language, which in turn will allow the creation of applications in various economic and social areas. BgGPT – the first open language model adapted to the Bulgarian language and created for the needs of Bulgarian society, is the work of the Institute for Computer Sciences, Artificial Intelligence and Technologies (INSAIT). VgGRT is already used by many different companies and countries. They share that he can reach the performance of the ShGRT on the final task in the Bulgarian language. Bulgaria is becoming one of the few countries that has its own system, taught in its own language and intended for use by everyone – schoolchildren, students, companies to develop applications, and also institutions to make analyses, reports and forecasts.  Chat GPT has advantages and disadvantages. It can be used by companies for 24/7 communication with customers, for endless support, better user experience, diversity and multilingualism. Disadvantages – lack of human emotions and reactions and depending too much on technology, unmanageable generated content with human biases and spreading misinformation. However, we’ll see how the system will develop. There are cases concerning GPT and copyright of the works on which it has been trained. For instance, ‘New York Times’ sues ChatGPT creator OpenAI, Microsoft, for copyright infringement. The claim was to end the practice of using its published material to train chatbots. While copyright protection generally extends to the creative works of individuals, it becomes more complex when it comes to AI-generated content. Systems like ChatGPT learn from vast amounts of data and create works on that basis, determining ownership of them is a legal and ethical challenge. OpenAI acknowledges that users are responsible for what they create using its API and encourages them to comply with applicable laws and respect intellectual property rights. Because models like ChatGPT are also trained on copyrighted material, there is a possibility that the generated content may inadvertently resemble existing works. Having in mind all that, let’s recall a few good practices: • Respect the rights of third parties: Keep in mind that while AI can produce impressive results, there are limitations and ethical considerations to keep in mind. • Involve human users in the creation process: this can help ensure that generated content complies with legal and ethical norms. • Treat computer-generated works as original: by including unique elements in your AI-generated content, you reduce the likelihood of infringing existing copyright norms and create something truly innovative. Artificial intelligence in media Its impact on the media, journalists, and audiences is complex and multifaceted. AI as facilitator of content creation. Huge benefits of AI in Media and Entertainment Industry The future of AI in media is promising, with the potential for AI-powered virtual and augmented reality experiences to transform media consumption. These technologies offer immersive and interactive experiences that can significantly enhance user engagement. AI in media governance as an independent structure. Streamline processes and improve services, resulting in more efficient operations. Enhanced budget allocation insights, leading to increased revenue generation. Real-time marketing intelligence, enabling greater marketing agility. On the other hand, the implementation of AI in the media can create new vulnerabilities, multiply information disorders, as well as generate challenges relating to transparency. Furthermore, it also risks becoming the back door to increasing the media’s dependence on the large technological platforms. In-depth studies of the impact of AI on the content and institutional characteristics of media are needed before steps towards specific regulation are taken. The Media and Information Society Committee of the Council of Europe (CDMSI) has established a subcommittee of experts on the impacts of generative AI on freedom of expression (MSI-AI). In 2024 – 2025 experts will work on creating guidance on the different types of impacts of generative AI on freedom of speech as a human right. They will not be binding on European countries, but can create the necessary basis for relevant national policies that consistently reflect the principles of human rights protection. The AI regulation proposed by the European Commission should bring clarity to the use of AI by the media as well. Main topics concerning the media • the analysis of automated information content (credibility, quality of information, ease of reading, etc.) • new journalistic formats • the relationship between platforms and media in an increasingly technological environment • the impact of AI on the privacy and transparency of media content • regulatory and legal developments of AI • the ethical repercussions of the use of automated content • new professional profiles and new tasks under AI use • changes in the training of journalists and communicators • the impact of AI on the employment and working conditions of journalists • the reception of automated content by audiences and its social effects (polarization, pluralism, etc.) • new digital divides (gender, race, social class, etc.). With the artificial intelligence expansion, there is a need to master a new type of competences related to the more complex social relationships between humans and intelligent machines. Media and digital literacy will be complemented by literacy about the characteristics and role of artificial intelligence in our world. Links: https://botpress.com/blog/are-there-any-legal-or-copyright-concerns-when-using-chatgpt-generated-content https://www.bloomberglaw.com/external/document/XDDQ1PNK000000/copyrights-professional-perspective-copyright-chaos-legal-implic

Opportunities and risks of artificial intelligence Chat GPT technology Read More »

The use of artificial intelligence (AI) and its consequences – Discussion on AI today goes in many directions

While the world word for 2023 was “Goblin mode”, artificial intelligence (AI) with 45% of respondents’ opinions, „assemblage“   with 37% and timeshelter with 35% are the symbolic words for 2023 in Bulgaria. The goal of the campaign about “the word of the year” was to trace public opinion about the most relevant topics of the past year through linguistic preferences and attitudes. Apparently, the phrase AI connects Bulgarian to global tendencies. Its use among Bulgarian population has risen five times in comparison to 2022. In the broadest possible sense, artificial intelligence is a machine that’s able to learn, make decisions, and take action—even when it encounters a situation it has never come across before. In the narrowest possible sci-fi sense, many people believe that AI refers to robots and computers with human or super-human levels of intelligence and enough personality to act as a character. What we have now is sometimes called weak AI, narrow AI, or artificial narrow intelligence (ANI): all programmes that are trained to perform specific tasks but aren’t able to do everything. This still enables some pretty impressive uses. Apple’s Siri and Amazon’s Alexa are both fairly simple ANIs, but they can still respond to a wide number of requests. Currently, most AIs rely on a process called machine learning to develop the complex algorithms that constitute their ability to act intelligently. European Union and artificial intelligence As a part of its digital strategy, the EU wants to regulate artificial intelligence (AI) to ensure better conditions for the development and use of this innovative technology. AI can create many benefits, such as better healthcare; safer and cleaner transport; more efficient manufacturing; and cheaper and more sustainable energy. In April 2021, the European Commission proposed the first EU regulatory framework for AI. It says that AI systems that can be used in different applications are analysed and classified according to the risk they pose to users. The different risk levels will mean more or less regulation. Once approved, these will be the world’s first rules on AI. After GDPR such comprehensive document will we of global significance and a leading example. The new rules establish obligations for providers and users depending on the level of risk from artificial intelligence. While many AI systems pose minimal risk, they need to be assessed. Several legislative initiatives are also under consideration to address various aspects of AI, such as the AI Liability Directive and the revision of the Product Defect Liability Directive. The first project aims to prevent the emergence of fragmented national adaptations of AI liability rules, while the second modernizes already existing rules to respond to the new challenges of the digital age. The AI Regulation will be the world’s first comprehensive instrument in this area. After the General Data Protection Regulation (GDPR), such a comprehensive document will have global significance and lead by example. Council of Europe The Council of Europe, which is the oldest human rights’ protection organization in Europe and the aim of which is to achieve greater unity among its members to “protect and realize the ideals and principles which are their common heritage”, adopted the first international, legally binding treaty to guarantee European values in the creation and use of artificial intelligence (AI) systems. The provisions of this convention govern the life-cycle activities of artificial intelligence systems and ensure their full compliance with human rights, democracy and the rule of law. Strict rules are laid down for the development, deployment and application of such technologies, with the 46 member states committing to adopt appropriate legislative, administrative or other measures to bring them into operation. These measures shall be graded and differentiated as necessary in view of the severity and likelihood of adverse impacts on human rights, democracy and the rule of law throughout the life cycle of artificial intelligence systems.  With a view to fostering innovation while avoiding adverse human rights consequences, each country is urged to create a controlled environment for the development, experimentation and testing of AI systems under the supervision of relevant competent authorities. The Convention is open for signature by the member states of the Council of Europe, by the non-member states that participated in its drafting, and by the European Union. Instruments of ratification, acceptance or approval shall be deposited with the Secretary General of the Council of Europe.  To ensure consistent implementation by Parties, the Convention establishes a follow-up mechanism and provides for international cooperation. A conference of representatives of the parties to the Convention monitors the effective implementation and identification of any problems and results of any reservation or declaration made under the procedures. Other tasks are to facilitate the exchange of information on significant legal, political or technological developments that would contribute to the achievement of the objectives pursued, to assist, when necessary, in the friendly settlement of disputes related to the implementation of the Convention and to promote cooperation with relevant stakeholders on various aspects of implementation, including and through public hearings when appropriate. Links:  https://www.europarl.europa.eu/topics/en/article/20230601STO93804/eu-ai-act-first-regulation-on-artificial-intelligence https://search.coe.int/cm#{%22CoEObjectId%22:[%220900001680afb11f%22],%22sort%22:[%22CoEValidationDate%20Descending%22]}

The use of artificial intelligence (AI) and its consequences – Discussion on AI today goes in many directions Read More »

European Commission has published guidance on protecting children online under the Digital Services Act (DSA)

Following consultation, the European Commission has published guidance on protecting children online under the Digital Services Act. Alongside the guidance, it has also issued a prototype of an age verification app. The guidelines adopt a risk-based approach, like the DSA, recognising that online platforms may pose different types of risks to minors, depending on their nature, size, purpose and user base.  Amongst other things, it includes recommendations on: Modifying the platforms’ recommender systems to reduce the risk of children encountering harmful content or getting stuck in rabbit holes of specific content, including by empowering children to be more in control of their feeds. Empowering children to be able to block and mute any user and ensuring they can’t be added to groups without their explicit consent. This could help to prevent cyberbullying. Ensuring that children’s lack of commercial literacy is not exploited and that they are not exposed to commercial practices that may be manipulative, lead to unwanted spending or addictive behaviours, including certain virtual currencies or paid loot-boxes. Introducing measures to improve moderation and reporting tools, requiring prompt feedback, and minimum requirements for parental control tools. The verification app will be tested and further customised in collaboration with member states, online platforms and end-users. The guidance outlines when and how platforms should check the age of their users. It recommends age verification for adult content platforms and other platforms that pose high risks to the safety of minors. It specifies that age assurance methods should be accurate, reliable, robust, non-intrusive and non-discriminatory. Ofcom UK has opened an enforcement programme to monitor whether providers are complying with their children’s risk assessment and record-keeping duties under the Online Safety Act 2023. The EU’s guidance differs from the very detailed guidance provided by Ofcom. All user-to-user and search services were required to carry out, and keep a written record of, a children’s access assessment by 16 April 2025 and where a service is likely to be accessed by children, those services must also carry out a children’s risk assessment by 24 July 2025 which requires providers to assess the risks of their service and associated content being harmful to children. Ofcom’s rules for protecting children come into force on 25 July 2025. Link: https://digital-strategy.ec.europa.eu/en/library/commission-publishes-guidelines-protection-minors

European Commission has published guidance on protecting children online under the Digital Services Act (DSA) Read More »

Laws that accelerate digitization in the European Union

The Digital Services Act (DSA) and the Digital Market Act (DMA) is a vast package of rules that applies through the whole European Union (EU). They pursue two main goals: DSA contains rules that primarily concern online intermediaries and platforms such as online marketplaces, social networks, content-sharing platforms, app stores, and online travel and accommodation platforms. The specific norms for very large online platforms and search engines which have more than 45 million users per month in the EU must abide by the strictest obligations of the Act.  The Digital Markets Act (DMA) includes rules that govern gatekeeper online platforms. Gatekeeper platforms are digital platforms that function as bottlenecks between businesses and consumers for important digital services. Some of these services are also covered in the Digital Services Act, but for different reasons and with different types of provisions. The Digital Services Act was adopted on 19 October 2022 came into force on 16 November 2022. Most services were given 15 months to comply with its provisions (until 17 February 2024). The Digital Markets Act entered into force on 1 November 2022 and on 2 May 2023, most of the DMA provisions became applicable. Under the Digital Services Act the European Commission has launched a Call for Evidence for guidelines on the protection of minors. The Commission uses Call for evidence to define the scope of a politically sensitive and/or important new law or policy, an evaluation of an existing law or policy, a fitness check of a bundle of related existing laws and/or policies. In order that online platforms comply with the DSA’s requirements to provide a “high level of privacy, safety and security for minors”, to ensure a harmonised implementation of the rules in all EU countries, and to help supervision and enforcement of the new regime, the Commission will draft special guidelines. To that end, the Call for Evidence invites submissions on the Commission’s proposed approach to develop good practices that platforms should adopt, including considering factors such as content moderation systems, commercial practices, and systems for selecting and presenting advertising. Input to the Call for Evidence will inform the Commission’s drafting of the guidelines on the protection of minors online, which will be opened to public consultation next year. Links: https://digital-strategy.ec.europa.eu/en/policies/digital-services-act-package

Laws that accelerate digitization in the European Union Read More »