Publication date: June 18, 2026
On June 19, 2026, representatives of our law firm will participate in the Data Science Summit AI Edition 2026, which will be held at the Palace of Culture and Science in Warsaw. For many years, the event has been one of the most important technology conferences in Central and Eastern Europe, bringing together specialists responsible for the development and implementation of the most advanced solutions based on artificial intelligence, machine learning, and data analysis. It provides a meeting place for the business, technology, scientific, and administrative communities—a space where implementation practice meets regulatory and strategic reflection.
Link to the event: https://ml.dssconf.pl/#agenda
This year’s edition focuses on the most important directions of development of modern artificial intelligence, including, among others, generative AI ( Generative AI), large language models (LLM), GPT systems, AI agents, Agent-to-Agent (A2A) architectures, Context Model Protocol (MCP), reinforcement learning, computer vision, predictive analytics, MLOps, LLMOps, AI governance and broadly understood data-driven business transformation.
The conference is aimed at data scientists, AI experts, data engineers, IT system architects, programmers, cybersecurity specialists, digital transformation leaders, product development managers, public administration representatives, and executives of enterprises implementing AI-based solutions. Speakers will include representatives of global technology companies, international corporations, financial institutions, telecommunications operators, energy companies, research and development organizations, academia, and technology startups – including AI directors, chief system architects, digital transformation leaders, scientists, AI engineers, MLOps experts, cybersecurity specialists, and managers responsible for implementing technological innovations.
The event is distinguished by its practical nature. The program focuses not only on presenting technological innovations, but primarily on showcasing real-world implementations already in place in businesses, public administration, the financial and energy sectors, e-commerce, telecommunications, healthcare, and industry.
Enterprise AI – From Experiment to Infrastructure
One of the most important themes of the conference is the practical use of artificial intelligence in production environments. It is increasingly clear that AI has ceased to be an experimental technology and has become a component of real-world business infrastructure. In this context, we are analyzing ways to counteract the phenomenon of so-called information bubbles and build more sustainable recommendation systems used by the largest media platforms. Methods are emerging to increase the diversity of presented content, reduce algorithmic bias, and design systems that responsibly influence user behavior.
A key issue is the collaboration between classic machine learning and generative AI. Combining both approaches allows for the development of scalable, cost-effective solutions, ready for use by large organizations. A significant portion of the discussion also focuses on countering financial fraud – methods for detecting fraud and building competitive advantage through structured AI system development processes. Developer experience and modern work environments for AI teams are also gaining importance, including ways to accelerate the creation and maintenance of AI-based applications.
The technological landscape is strongly influenced by the prototyping of AI agents using modern programming environments and platforms supporting the creation of autonomous systems. Simultaneously, there are emerging themes regarding the use of AI in the legal sector – methods for assessing the reasoning ability of language models, methods for tracing legal sources, and techniques for increasing the reliability of AI-generated responses. From a security perspective, mechanisms for securing autonomous systems against attacks, unauthorized access, and attempts to manipulate their behavior are particularly important.
Equally important is the optimization of language models – techniques for quantization, distillation, and model size reduction while maintaining high efficiency. In the financial sector, solutions are emerging that include automated document processing, data extraction, and the development of intelligent analytical platforms. As an illustration of the creative potential of AI, the mechanisms behind music generation by artificial intelligence and the application of deep learning in creative processes are also discussed.
Generative AI – Multimodality, Agents, and Public Administration
Generative AI is one of the fastest-growing areas of contemporary technology. In production practice, it is used to automatically describe works of art, create alternative descriptions that increase the accessibility of digital documents for people with disabilities, and develop strategies for replacing very large and expensive language models with smaller, specialized solutions that can be effectively used in enterprises. The evolution of architectures leads from simple prompts, through RAG mechanisms, to multi-stage pipelines integrating multiple data sources and advanced LLM architectures.
Multimodality plays a significant role – solutions that can simultaneously analyze text, images, sound, and spatial information – opening the door to building more versatile AI systems. In enterprise environments where standard communication protocols aren’t feasible, building agents that exclusively utilize approved APIs, internal tools, and development platforms becomes crucial, while adhering to security and governance principles.
A crucial direction is the use of artificial intelligence in public administration, including the process of transforming an AI project into a product accessible to millions of citizens. Solutions that generate sound, speech, and music using modern diffusion models and transformers are also gaining importance, as are automatic analysis of PDF documents and the extraction of structured data from multi-page administrative materials. Hybrid pipelines combining generative AI, signal processing, and embedded systems for use in resource-constrained devices are also emerging. A separate, increasingly important issue is assessing the quality of RAG chatbots, testing them before deployment, and building reliable conversational systems.
Machine learning in practice – recommendations, energy, finance
Classic machine learning remains relevant – on the contrary, it’s finding applications in major industrial and financial implementations. Modern recommendation systems are emerging, based on data representation vectors and embedding architectures, where balancing recommendation effectiveness with the ability for users to discover new content is paramount. Digital twins are being used to optimize energy assets and manage industrial infrastructure. In the financial sector, teams of collaborative models are emerging to support decision-making processes.
The challenges associated with AI agents operating in social media are increasingly being discussed – issues such as user privacy, data security, and customer interaction design. In the banking sector, there’s a clear trend toward building millions of personalized models operating in parallel, which requires advanced training optimization, system scaling, and efficient processing of massive data sets.
Deployments, MLOps, and Security – The AI Systems Lifecycle
One of the most practical areas of the event is implementing, maintaining, and scaling AI systems in production environments. Simply developing a model is no longer the biggest challenge; ensuring its stable, secure, and effective operation in real-world business environments is crucial.
The security of generative AI is becoming one of the most pressing issues. Practical methods are emerging to protect models from prompts, injection, jailbreaking, and attempts to bypass control mechanisms. Referring to OWASP guidelines and modern solutions enabling the implementation of AI systems in accordance with the security- by-design principle is becoming standard.
From a legal perspective, issues related to designing global data platforms operating simultaneously in multiple jurisdictions are particularly important. These challenges include data sovereignty, local regulations, and the need to reconcile legal requirements across different jurisdictions.
A key theme is the full lifecycle of machine learning models – the processes of building, deploying, monitoring, updating, and retiring AI models, as well as how to manage their performance at scale. System performance optimization plays a significant role – techniques for reducing latency, shortening inference times, and adapting models to various hardware environments. Case studies from the pharmaceutical sector also appear, demonstrating how to manage chatbots’ deviance from their original purpose and how to minimize the risk of generating undesirable responses. Simultaneously, computer model optimization techniques are being developed, allowing to increase their efficiency many times while maintaining high quality of prediction.
Monitoring LLM systems and AI agents before end-user errors occur is becoming particularly important. Systems for model tracking, automated evaluation, anomaly detection, cost management, and continuous response quality improvement are emerging. Solutions are emerging that enable the management of thousands of AI agents and their evaluation at scale. This is complemented by approaches based on GitMLOps, DevSecOps, and modern observability solutions for systems based on large language models.
AI Agents – Autonomous Decision-Making Systems
One of the most important areas of the conference is AI agents. Modern artificial intelligence is increasingly moving beyond the role of a mere tool for human support and instead becoming an autonomous system capable of executing complex business processes. Methods are emerging for assessing the effectiveness of large-scale multi-agent systems – methods for automated testing, reliability testing, tracing error sources, and building systems that guarantee repeatable results. Significant attention is being paid to monitoring agents in real time, identifying errors, measuring response quality, and continuously improving decision-making processes.
Issues related to the security of AI agents are becoming increasingly important, including attack scenarios that confuse autonomous systems and methods for securing them against manipulation. Practical applications of agents utilizing financial, registry, and business data are emerging, supporting analytical, scoring, and decision-making processes. AI agents are also finding applications in telecommunications and in the construction of systems that independently conduct research and analysis.
Agents’ long-term memory plays a significant role – solutions utilizing knowledge graphs and mechanisms that enable agents to independently organize and expand their accumulated knowledge. Real-time voice agents are highly practical – systems capable of conducting telephone conversations with users, analyzing speech, and generating responses that approximate natural communication. Solutions are also emerging that enable automatic detection of problems arising during the operation of production agents and the generation of evaluation processes without human intervention. Applications also include systems supporting energy management, intelligent buildings, and Internet of Things devices. A separate, important area is the evaluation of Text -to-SQL systems, which enable the automatic conversion of natural language into database queries.
Development and tuning of large language models
A significant portion of the conference was devoted to the creation, training, and refinement of large-scale language models. The most common errors made when building AI models are increasingly understood, as is the significant impact that the proper selection of training data has on their quality. Specialized models for the banking sector are being developed, along with methods for adapting them to specific business applications.
A significant area of research is Polish multimodal models capable of simultaneously understanding text and images, including methods for building them with limited data resources. Techniques for simplifying official and administrative language using language models are also emerging, with a particular emphasis on increasing the accessibility of public transportation. Mechanisms for building trust in artificial intelligence by providing users with justifications for generated responses are gaining increasing importance. Embedding models, used to identify duplicate data, ambiguous names, and unstructured addresses, are becoming increasingly important in everyday business practice.
Business Transformation – Strategy, Data, and Organization
From a strategic perspective, artificial intelligence is becoming one of the most important elements in building a competitive advantage for enterprises. Practical experience is emerging with implementing generative AI in international organizations and managing technological transformation processes. Methods for identifying the most promising AI initiatives from among hundreds of potential projects are becoming increasingly important.
Data quality plays a significant role – proper data preparation is now a prerequisite for the effective implementation of artificial intelligence. Critical analysis of technological trends is also emerging, allowing us to separate the true capabilities of AI from marketing hype. The importance of data architecture and IT systems for the success of transformation projects is becoming increasingly clear. The debate on the future of management in the era of agent-based AI and the changes AI will bring to corporate organizational structures is particularly prominent. Examples of building scalable customer service agents operating simultaneously in multiple countries are also emerging, as are the experiences of large organizations implementing AI, developing data products, supporting business expansion, and transforming the banking sector.
AI in Software Development – Regulation, Security, and Intellectual Property
Artificial intelligence is fundamentally changing the way technology teams work. Building local systems to support programming in compliance with European regulations, including the AI Act and NIS2, is becoming increasingly important. Both the potential and limitations of modern programming assistants are emerging, including the reasons for the failure of AI agents working with enterprise data and ways to build more predictable systems.
Cybersecurity, threat modeling, and vulnerability management in AI-based systems are all important topics, as are the practical aspects of building agent platforms capable of operating in production environments. From a legal perspective, issues related to intellectual property protection, commercialization of AI-based solutions, copyright management, and building secure technological product development processes are particularly important.
Computer Vision – intelligent image analysis
Modern image analysis goes far beyond classical object recognition. New approaches are emerging to using visual models in unusual applications, including Visual RAG systems that enable automated processing of data contained in spreadsheets and corporate documents. The technological trade-offs involved in designing high-performance pipelines are increasingly understood. computer vision. Computer vision is also used in climate forecasting and analysis of environmental phenomena, as well as in automatic data extraction from Polish administrative and business documents.
Debates about the future of artificial intelligence
An integral part of the event are expert roundtable discussions. The role of software agents is discussed – whether they will become merely a tool to support specialists or will they take over some of their responsibilities. The use of AI agents as digital equivalents of junior engineers supporting data science teams is also discussed. Experts engage in debates on the labeling of AI-generated content and the transparency of AI systems.
A separate discussion is devoted to Europe’s technological sovereignty and the possibility of building an independent artificial intelligence ecosystem. The discussion concludes with a discussion on the competencies of the future and the skills that will become crucial for organizations to function in an AI-based economy.
The Data Science Summit AI Edition 2026 demonstrates that artificial intelligence is no longer just a technology of the future. It has become one of the most important tools shaping modern businesses, public administration, and the financial, energy, industrial, and technology sectors. The scale and scope of the topics covered confirm that AI development today requires not only advanced technological competencies but also appropriate risk management, security, regulatory compliance, and a responsible approach to implementing new technologies – making law firms’ participation in such events a natural part of tracking the directions in which new technologies law is heading.
#DataScienceSummit #AIEdition2026 #ArtificialIntelligence #AI #MachineLearning #GenerativeAI #LLM #GPT #AIAgents #AgenticAI #AgentToAgent #ModelContextProtocol #MCP #DeepLearning #ReinforcementLearning #ComputerVision #PredictiveAnalytics #MLOps #LLMOps #AIGovernance #ResponsibleAI #AIEthics #AISecurity #PromptInjection #Jailbreaking #SecurityByDesign #OWASP #DevSecOps #GitMLOps #CyberSecurity #DataSovereignty #AIAct #NIS2 #EURegulations #TechSovereignty #DigitalTransformation #BusinessTransformation #DataDriven #DataQuality #DataEngineering #DataScience #CloudComputing #RAG #VisualRAG #Multimodal #Multimodality #Embeddings #VectorSearch #Quantization #ModelDistillation #FineTuning #TextToSQL #KnowledgeGraphs #VoiceAgents #ConversationalAI #Chatbots #DigitalTwins #IoT #SmartBuildings #EnergyManagement #FinTech #BankingAI #LegalTech #LawFirm #IntellectualProperty #Copyright #PublicAdministration #GovTech #HealthcareAI #PharmaAI #Telecommunications #SoftwareDevelopment #AIAssistants #CodingAgents #Innovation #TechConference #Warsaw #CEE #KGLegal #KieltykaGladkowski