AI is no longer a distant promise; it is the new language of global competitiveness. For Italy, this represents a unique opportunity: a landscape of SMEs and industrial districts where creativity and craft meet concrete value. The challenge now is to move beyond isolated experiments and transition toward true industrial capability. We need a "factory" of AI results—repeatable, scalable, and human-centric. Marco Fanizzi, VP & Managing Director of Dell Technologies Italy, explores the "Dell AI Factory" vision: a holistic ecosystem of infrastructure, data, and services designed to bring AI to the heart of decision-making—from the edge to the cloud. By championing an open ecosystem over closed stacks, Dell is empowering organizations to lead in the "Agentic Era." This isn't about replacing Italian talent; it’s about amplifying it—making it faster, more inclusive, and more competitive on the world stage.

The AI revolution compels us to rethink how we interpret constitutional principles in the digital society—such as freedom, equality, dignity, and accountability. The challenge is to move beyond a logic of mere compliance: what is needed is adaptive AI governance, capable of evolving alongside the models themselves. Companies are no longer merely recipients of rules, but co-creators of a digital constitutional ecosystem. The contribution puts forward a new “constitutionalization of innovation.”

Artificial Intelligence is the cornerstone of ESA’s strategy for the next frontier of Earth Observation. At the heart of this mission is Φ-lab: an innovation hub designed to bridge the gap between research, industry, and bold new ideas. We shepherd transformative concepts through the entire innovation lifecycle—turning early-stage experiments into high-impact global solutions.

Like for any new technology, there are different ways to build AI agents. Whether we choose to rely on big LLMs, commercial black-box AI tools or smaller, open-source models --  every small choice we make along the way can have huge repercussions. In my talk, I will address how to find the right balance through informed choices, resulting in cutting-edge AI technology that is reliable, cost-efficient, and sustainable.

The proliferation of Artificial Intelligence does not signal the displacement of the human, as many fear, but rather the emergence of a “hyperhuman” dimension. In this new paradigm, human and artificial intelligence converge to augment the processes of knowledge creation and learning. However, AI—particularly Large Language Models—is inseparable from large-scale computational power. While High-Performance Computing (HPC) provides the material foundation for training and deploying complex models, viewing it merely as "raw power" is reductive. In the modern era, computing infrastructures are cultural infrastructures: they dictate which projects are viable, which datasets are prioritized, and who has the agency to innovate.
Within the Italian landscape, CINECA serves as a definitive case study. A non-profit inter-university consortium overseen by the Ministry of University and Research and the Ministry of Education and Merit, CINECA operates at the intersection of world-class HPC infrastructure and digital service provision for the public sector. This dual mission—advancing supercomputing while developing academic information systems—offers a tangible look at the deep interplay between computational backbone and digital services for research, teaching, and advanced training.

AI is no longer limited by algorithms—it is constrained by energy, latency, materials, and trust. In this keynote, Rika Nakazawa, executive leader at NTT DATA, reframes the AI debate from model supremacy to system design at planetary scale. Drawing on NTT DATA’s work across AI platforms, photonic networks, and emerging quantum infrastructure, alongside her applied doctoral research, Nakazawa introduces Organic Intelligence: a design paradigm that treats intelligence as embedded in physical infrastructure, energy flows, and human context. Moving “from code to carbon,” she explores how bio-inspired systems, optical networks, and observation-driven AI can deliver scalable performance while dramatically reducing environmental and operational cost. This keynote positions AI not as a race for raw compute, but as an engineering challenge of integration, resilience, and alignment—where leaders capable of designing end-to-end intelligent systems, from infrastructure to impact, will define the next era of trusted AI at scale.

Historically, theatre has always been a space for experimentation—pushing the limits of narrative, presence, and reality. Today, with the rise of Artificial Intelligence, it can reclaim that role once again. Between generative texts, glitched identities, and digital rebellions, theatre transforms into a space where the author is an algorithm, the actor is a fragment of code, and the audience participates in unpredictable endings. This is a performative reflection on authorship, identity, and the future of storytelling in the age of creative machines. Ultimately, it points to a new frontier for entertainment, unlocking immersive, interactive experiences that redefine how stories are told, shared, and lived.

AI is a general-purpose technology, and gaining real competitive advantage means moving beyond isolated experiments to structural reinvention. This keynote explores how leading organizations redesign workflows, operating models, and value creation around AI. You’ll learn practical principles to move from vision to execution, accelerate transformation in 90 days, and build the governance and systems needed to scale AI safely and sustainably.

In this talk, we will showcase Anthropos’ AI Simulations solution, sharing the insights and data gathered from analyzing over 2,000 candidates and nearly two years of developing our technology. The session aims to show managers, HR professionals, and entrepreneurs how AI-driven skills assessment can streamline business processes—from recruiting to training—and deliver immediate ROI in terms of employee productivity and their ability to transition into new roles.At the end of the talk, we will also share our journey in using AI to build our technology, the lessons learned, and practical advice.

The talk explores AI safety challenges—hallucination and prompt injection—through multi-agent architectures and open-source standards. It presents the Open-Floor Protocol (Linux Foundation AI & Data) and new frameworks for hallucination and injection mitigation, using iterative review, fact-checking, and security KPIs to enhance reliability, transparency, and trust in generative AI systems.

Imagine a car driving through the city and, thanks to artificial intelligence, detecting in a single pass all kinds of road and sidewalk anomalies — the things we normally miss: faded markings, missing signs, drains, potholes, cracks, overgrown vegetation. It works on any type of road at speeds up to 130 km/h and uses up to three cameras. One single pass generates a priority map with 20 cm accuracy and authenticated data, turning observation into rapid, concrete action.

The roundtable will allow us to explore how to build trust in intelligent and autonomous systems. In a landscape driven by innovation and increasing automation, with the rise of Agentic AI, trust becomes the central resource: how can we ensure transparency, reliability, and control when algorithms no longer merely suggest but begin to act? What new risks emerge, and what responsibilities must institutions, developers, and users take on?

Artificial intelligence is profoundly transforming every economic sector, and healthcare is no exception. Its adoption is rapidly accelerating, with applications ranging from biomedical research to the optimization of administrative and management processes, all the way to supporting clinical activities. Scientific evidence helps us address key questions: where do we stand today? What lies ahead? Will AI expand access to care, or will it become yet another factor of inequality in an Italian healthcare system that already shows significant disparities in the actual fulfillment of the right to health?

Introducing AI is not enough — organizations must rethink models, workflows, skills, and leadership. What they need is an AI readiness strategy. The session covers the five key levers of AI readiness (processes, data, mindset, governance, upskilling), concrete case studies drawn from real-world experience, and practical tools such as AI activity mapping, maturity assessments, and controlled experimentation. It also highlights the crucial role of managers as facilitators of change, offering a clear and actionable framework to help organizations transition effectively toward an AI-powered future.

Artificial intelligence through voice is transforming Service Management. Responsa’s Voice Agent listens, understands, and acts: it streamlines processes, enhances corporate knowledge, and creates an ecosystem of intelligent collaboration between people and tools.

The panel explores how new entrepreneurial initiatives are created, tested and scaled. Through real-world cases and direct experiences, it will examine how agentic AI is transforming insight generation, concept validation, team building and the day-to-day operations of early-stage startups. A practical discussion on the opportunities, limitations and organizational impacts of this new era of venture creation.

At the heart of today’s democratic debate, artificial intelligence is becoming a pivotal force in persuasion, consensus-building, and the shaping of the public sphere. This panel explores how the algorithmic world is redefining not only political dynamics but also the formation of new generations—particularly university students, who are increasingly exposed, involved, and required to interpret an information ecosystem that has been profoundly transformed. The talk will examine the impact of intelligent systems in shaping opinions, identities, and participatory behaviors, with a focus on those entering the professional world today and those who will do so in the coming years. Through case studies, recent research, and interdisciplinary insights, the discussion will investigate risks, opportunities, and emerging responsibilities. In an era of automated communication, profiling, and sentiment analysis, a central question arises: are we, as individuals and as a society, ready to understand and govern the power of the systems that shape what we see, think, and share? A panel that brings together technology, politics, critical education, and democratic responsibility, offering tools to interpret—rather than merely endure—the new horizon of the AI Generation.

A conversation to understand how to balance human intuition and algorithmic power in a new era of investing. The panel will explore how AI can become a true strategic partner for investors, enabling faster and more accurate data-driven decisions.

We are entering an era in which AI and people collaborate naturally, creating intelligent CX ecosystems that transform the customer experience. By combining Agentic AI with a structured information asset, it becomes possible to orchestrate more effective, seamless, and personalized interactions. Through real demonstrations, we will show how knowledge, context, and purpose can become decisive levers for creating surprising experiences that are truly value-driven.

We are at a turning point in AI content generation. With the UNet era now behind us, the architectural debate between single-stream and dual-stream DiT models is reshaping the field. On one side, Qwen’s dual-stream MMDiT architecture—implemented in WAN Video—processes text, video, and spatial controls through separate streams that interact via joint attention, enabling granular control over camera, motion, and frame-by-frame editing. On the other side, Z-Image’s single-stream S3-DiT shows that the “scale-at-all-costs” paradigm can be challenged: just 6B parameters compared to the proprietary giants’ 20–80B, yet delivering competitive performance and sub-second inference on consumer GPUs. In this panel, we will explore key questions. Why does dual-stream MMDiT unlock motion-transfer and spatial-control capabilities that were previously unreachable? How does the single-stream S3-DiT achieve such high efficiency while maintaining SOTA quality? Big Tech vs. Open Source: who is actually winning when democratization through LoRA and quantization competes with pure computational power? And what limitations remain unsolved? Temporal consistency beyond 15 seconds, physics-aware generation, memory footprint, and energy efficiency are still open challenges for both approaches. A technical deep dive into how different architectural innovations are rewriting the rules of AI generation.

In this breakout session co-presented by Dell and Intel we will get into the concrete of how to build a true enterprise, open and scalable AI platform, starting from field experience with customers from different industries.​ We will be showcasing the Dell AI Factory, Dell's end-to-end approach that combines AI- optimized infrastructure, data platforms, specialized services and an open ecosystem of partners and technologies to bring AI as close to data as possible, from the edge to the data center and the cloud. We'll focus specifically on the Dell AI Platform with Intel, an end-to-end validated solution with an open-source software stack designed for generative AI and machine learning workloads at scale. 

The emergence of autonomous AI agents marks a paradigm shift: for the first time, artificial systems can make decisions and act without direct human supervision. This autonomy raises unprecedented legal and operational questions that every startup must address to mitigate real risks.This panel explores the practical challenges of implementing autonomous agents. We’re not talking about technological hype, but about how to build systems that are safe, traceable, and compliant when the rules of the game are still emerging.We will tackle three critical questions. When an AI agent makes an improper decision, who is legally responsible? How can full traceability of actions be ensured for agents operating in real time, making hundreds of decisions per day? What kill switch mechanisms allow stopping a misbehaving agent while balancing autonomy and safety?

How does industry change when computer vision meets generative AI? We will explore how Vision Language Models (VLMs) are becoming practical tools for industry. Multimodal models that can see, understand, and interact with diagrams, images, drawings, and technical texts, supporting quality control and the optimization of production processes. A rapid leap from academic innovation to real-world solutions for production lines and complex processes.

 In an era where artificial intelligence is reshaping industries at exponential speed, leadership must evolve just as rapidly. This keynote explores how CEOs and executive leaders can navigate the AI age—not through more data or automation alone, but by embracing what Dr. Tina Ruseva calls Soft Work: the human-centric leadership competencies that remain irreplaceable in a machine-driven world. From ethical decision-making and emotional intelligence to creativity, purpose, and trust-building, Soft Work is the new strategic advantage. Drawing on real-life insights from leading an AI startup and years of future-of-work research, this talk challenges traditional leadership models and offers a bold framework for adaptive, humane, and effective leadership in a time of intelligent disruption. Audiences will leave with: A clear understanding of how AI is transforming the CEO role A practical toolkit for cultivating future-ready leadership A redefinition of intelligence in business—one that puts humans back at the center This keynote is a call to action: in the age of artificial intelligence, it’s time to lead with deeper human intelligence.

TEDA Tirana is Albania’s flagship free economic zone, located just 10 minutes from both Tirana city center and the International Airport, with direct access to the Port of Durrës and Corridor VIII. Investors benefit from significant tax and customs incentives, long-term lease agreements starting from €1/m², and fully operational logistical infrastructure. A key competitive advantage is energy cost efficiency: over 95% of Albania's energy consumption is generated from clean sources, significantly reducing energy-related expenditures. TEDA Tirana is the ideal location for AI, R&D, ICT, and manufacturing companies seeking fast, secure, and competitive access to European markets.

In the agentic era, competitive advantage is no longer about doing more, but about doing it first. Rapidly transforming ideas into products, services, and new markets has become the true frontline of competition. AI agents are enabling this shift, redesigning the way companies make decisions, experiment, and innovate. It is no longer about merely optimizing existing processes, but about accelerating the capacity to act. While data remains fundamental, it is not enough on its own. Value emerges when Artificial Intelligence understands the context in which it operates—goals, processes, people, and business dynamics—and uses this understanding to drive relevant decisions. From SMEs to large enterprises, the agentic era is compressing time-to-market, making ideas scalable and unlocking new business models. Competitive advantage belongs to those who can transform intelligence into action at the right moment.

Replace guesswork with testable decisions. Agentic AI now lets teams generate digital populations that mirror their markets. This allows you to preview how customers react to onboarding flows, fee disclosures, APR moves, cross-sell prompts, and collections outreach. Model churn before a price change. See which disclosures lower complaints. All tests run in a sandbox with policy rules enforced. I will show how teams shipped only what worked and cut time-to-market.

Cybersecurity is the latest revolution in warfare, reshaping the fate of the world. GenAI dominates the battlefield and is used by attackers for sophisticated, automated attacks—from hyper-personalized phishing (deepfakes) to adaptive malware. Traditional signature-based defense is conceptually outdated. We explore how AI, in partnership with human intelligence and platforms such as EDR/XDR and SIEM/SOAR, is enabling autonomous defenses to counter advanced adversaries.

AI is not just a ‘technology add-on’, but a paradigm shift that demands careful design. With an approach that combines strategic design and AI, we will share tools and principles to understand usage contexts, shape intelligent experiences that foster adoption, and prototype more rapidly to bring the product vision to life.

The introduction of GraphRAG, hybrid RAG+KG systems, and vertical LLMs/SLMs for clinical domains brings new structural challenges: the lack of gender-disaggregated data, biased graph nodes, and embeddings that fail to represent minority populations. This session offers a technical discussion: How does dataset unevenness affect retrieval and reasoning? Which metrics are needed to evaluate gender misrepresentation in RAG systems? And how can we model dynamic, geolocated queries in low-information-density contexts? I will share insights from Geen, a GraphRAG system for medical triage, to explore failure cases, performance limitations, and retraining strategies. A conversation designed for engineers, developers, and knowledge-system architects.

How do you build a competitive advantage when AI becomes an autonomous agent? A practical journey into AI that supports both those who design the future (R&D) and those who manage the present (Operations). Together with B4I – Bocconi for Innovation, we explore two crucial stages of the corporate value chain. QGrain, the platform that reduces risks and R&D timelines by providing market intelligence and automated compliance checks for new products. ProcederAI, which transforms operations by converting videos and actions into standardized procedures and interactive training.

Generative AI is rapidly entering higher education, yet many universities struggle to translate ethics, policy, and technical potential into meaningful learning practices. Based on insights from the ADMIT European partnership, this talk explores how institutions are embedding responsible GenAI into learning design, assessing organisational AI maturity, and supporting educators through practical frameworks. The focus is on when GenAI adds real educational value, and when restraint is needed. Not a slide-only talk, expect moments of interaction and collective sense-making.

The presentation analyzes the evolution of artificial intelligence in healthcare, from decision support systems to GenAI, up to the most recent Agentic AI approaches. Through use cases and ongoing experiments, the speech discusses opportunities and limits of adoption in the healthcare field, highlighting the main organizational and managerial implications. Attention is placed on the conditions that favor the transition from experimentation to implementation.

Autonomous agents are transforming how startups grow: commercial pipelines, R&D, cybersecurity, on-demand content, automated governance and compliance. How to integrate agentic systems into core processes: from content and threat analysis to the creation of personalized micro-courses and the automation of strategic data analysis. A panel designed for those who want to push their startup into a new phase of efficiency, execution speed and operational scalability through advanced agentic systems. Join a practical, founder-to-founder discussion on: which processes can already be agentized today (and which will be in the next 6–12 months); what it takes to design and orchestrate hybrid human + agent teams; how agents are reshaping media intelligence, cybersecurity, edtech and data-driven consulting; and which capabilities founders need to truly leverage Agentic AI.

What is the value of privacy, data security, and regulatory compliance in the era of AI? What we now call data sovereignty has become an essential requirement for many companies. This is where NT Nuove Tecnologie comes into play, together with NVIDIA technologies, to develop Private AI projects: complete hardware-and-software infrastructures — customized, integrable, and designed to create an alternative or complement to the most well-known open-source solutions.

The talk will provide an overview of what is meant today by Agentic AI and will examine its key characteristics within the evolutionary process that has been shaping process automation in recent years. Drawing on data from the Research of the Intelligent Business Process Automation Observatory of the Politecnico di Milano, it will also offer a snapshot of how Italian and European companies are approaching this revolution: which technologies, organizational choices, and projects are currently attracting the greatest interest.

Italy has taught the world to build with beauty, ingenuity, and vision. But in the era of artificial intelligence, “Made in Italy” still seems to be in a phase of catching up. And yet, the potential is all there: in our ability to unite creativity and method, craftsmanship and technology, intuition and rigor. “Made in Ital.I.A.” is an invitation to rethink artificial intelligence as an expression of our cultural DNA. A story of data and courage, which shows how Italy can lead the transformation not by copying the models of others, but by building its own: based on trust, governance, and humanistic intelligence. And perhaps the next step of “Made in Italy” will be exactly this: learning also to buy Italian, to trust our skills, to build together an ecosystem that makes AI truly ours — beautiful, solid, and above all, with an Italian soul.

AI is introducing a new operational paradigm across the food value chain: digital agents capable of making autonomous decisions, adapting to data, and coordinating complex processes in real time. From demand planning to quality management, from logistics to sustainability, this evolution promises to make the supply chain more efficient, resilient and transparent. The panel explores concrete application cases and the impact on producers, startups and major brands.

A journey into the organization to discover how AI becomes an infrastructure that connects processes and people. Two perspectives—company and individual—to understand how artificial intelligence makes work more effective and sustainable, fostering collaboration and innovation.

The AI agents revolution is not just a matter of advanced technology, but of accessibility and concrete value. IT4LIA AI Factory emerges as an essential catalyst for Italian SMEs, offering infrastructure, tools, and expertise to integrate intelligent agents into business processes. This talk explores how small and medium-sized enterprises can leverage the Factory's services—from rapid prototyping to implementation support—to compete in the era of agents, creating real and sustainable competitive advantages in the European digital market.

End-to-end testing of web applications is essential to ensure product quality and stability. Yet traditional approaches to automated E2E testing often clash with complex technologies, intricate user flows, long development timelines, and high maintenance overhead. But what if we could leverage the power of Large Language Models and AI agents to create automated tests that are robust, intelligent, and self-adapting? In this session, we explore the concrete technologies that enable the use of AI to automate E2E web testing. We’ll look at examples of intelligent agents capable of understanding interfaces, adapting to UI changes, and autonomously generating new test cases — ready to be unleashed into the testing environment in search of bugs.

A preview of the survey by the CNA Research and Studies Area on the relationship between Artificial Intelligence and Italian micro and small enterprises. The data shows a growing interest, with AI predominantly perceived as a tool for operational efficiency. The main barriers to adoption are not economic, but concern understanding ("understanding what is useful") and the lack of internal skills. The emerging support needs will be discussed, with an emphasis on the crucial role of training and guidance to fully integrate AI into the productive fabric.

From cloud to mission: how AI agents are evolving from connected assistants to autonomous intelligences, operating in physical and immersive environments. In the most critical contexts, on-premise is a strategic choice for security: data under control, lower exposure, operational continuity, and compliance. Concrete examples and practical insights to evaluate the most suitable architecture.

Small and medium enterprises make up over 90% of the Italian production ecosystem, acting as the vital core of industrial districts and local economies. This roundtable explores how, with the introduction of Agentic AI, these companies can now access technologies that were once reserved for large enterprises due to the significant investments and dedicated teams required.

How are Artificial Intelligence technologies transforming the analysis, preservation, and enhancement of Cultural Heritage and archaeological practices? The talk presents real-world use cases—from automated identification to digital reconstruction and robotics—highlighting opportunities, limitations, and future perspectives for the field.

[In-person workshop with remote speaker]. Got an idea but intimidated by the code? Then it’s time to dive into "Vibe Prototyping"! In just 75 minutes, we’ll turn your words into reality. We will use Google AI Studio to transform your intuition into a concrete project through prompt engineering, and then leverage the magic of Firebase Studio to generate the app instantly. This is a technical workshop designed for everyone: whether you’re a Developer or a PM, a Designer or a Manager, all you need to bring is your creativity (and a Google account, if you like). From prompt to working prototype at the speed of light! This workshop is reserved for holders of an AI Festival 2026 ticket and also requires a dedicated ticket, available for separate purchase, to access the individual session.

By definition, Artificial Intelligence is a macro-scale phenomenon, favoring vast datasets, massive computational power, and significant capital concentration. This raises a critical question: is the Italian model of 'micro-efficiency'—typified by the visionary entrepreneur, the agile SME, and the exit-driven startup—still viable in this new era?

An interaction engineer studies how to make digital technologies more intuitive and sustainable for humans. In this session, he will explain why it is no longer useful to view artificial intelligence merely as a tool, but rather as an interlocutor we communicate and collaborate with—one we can learn from and, in many situations, rely on, both at work and in everyday life. This relationship of reliance can be appropriate or not, both cognitively and ethically, and it must remain sustainable on a human level as well, so as not to erode our knowledge base and our motivation. The goal is to create hybrid intelligence: a form of human intelligence—ours—enhanced by technology. This hybrid intelligence emerges from a functional, conscious, and responsible collaboration between humans and AI systems: an AI that makes us more capable, not more dependent.

We are witnessing a wave of "hyper-innovation" driven by AI, yet its success isn't plug-and-play. To move beyond the buzzwords, we must overhaul the frameworks surrounding the technology—including policy, academia, and social infrastructure. This talk challenges the mainstream narrative, arguing that AI is a catalyst that requires systemic evolution to be effective. Drawing on historical parallels from the history of science, we will explore why technology alone is never enough to change the world.

Artificial Intelligence doesn’t replace the discipline of data analysis; it amplifies it. In this workshop, we will explore how to integrate AI across the entire data-driven workflow—from measurement to predictive modeling, and from insight extraction to data storytelling—while keeping core analytical skills at the center. AI is more than just a tool for faster data visualization: it can become a true engine for building custom data analysis systems capable of transforming raw information into actionable decisions. During this session, I will share the framework I developed to build complete, standalone analysis systems—independent of traditional AI interfaces but designed using AI strategies and techniques. Together, we will work step-by-step on your own dataset (or one provided by me) to reach the final goal: publishing your custom analysis system. No development skills are required. I am not a developer myself, and this method is specifically designed for professionals who want to build functional prototypes by maximizing AI’s potential. Workshop Outcomes: 1 Complete & Customized Data Analysis System: Built during the session; 1 Replicable Framework: For use in all your future projects; 1 Structured Method: To apply AI to analytics with a professional, systematic approach. This workshop is reserved for holders of an AI Festival 2026 ticket and also requires a dedicated ticket, available for separate purchase, to access the individual session.

In recent years, the AI discourse has been dominated by the models themselves. Today, however, a critical shift is occurring: competitive advantage no longer stems from having the "best" model, but from the ability to integrate intelligence into functional, operational systems. This session explores the evolution from LLMs-as-APIs to agentic AI systems capable of understanding and acting within the real world. We will argue that we are not witnessing a model revolution, but a systemic one. Through real-world production use cases, we will examine what it means to design AI-first solutions, build robust application layers, and move AI beyond the "demo phase" into reliable, large-scale implementation. Finally, we will look toward the future—exploring exponential growth loops and world models—to understand how intelligent systems are evolving and how to prepare for what comes next.

AI isn’t replacing the workforce, it’s empowering it. Join us as we dive into how forward-thinking organizations are rewriting the rulebook on roles and operations to unlock real-world results. We will break down successful human-machine partnerships and share a roadmap for closing the skills gap, ensuring your talent is ready to lead a new wave of productivity that transcends automation.

An inside look at the final weeks before ChatGPT’s public debut. Through behind-the-scenes stories and exclusive anecdotes, we revisit the dawn of OpenAI’s chatbot—the exact moment AI evolved from raw computing power into an unexpected persona, sparking the fastest-growing online service in history.

The talk explores how to design AI-ready business processes by adopting a Compliance by Design approach. It will cover governance, impact assessments, risk management, and the requirements of the AI Act’s risk-based framework. The focus will be on integrating these elements into organizational models, controls, accountability, and privacy protection to enable responsible and effective AI adoption.

Artificial intelligence promises great improvements, allowing us to exploit enormous volumes of data to generate predictions and automate workflows and processes. However, a passive trust in systems whose logic is not clear to the decision-maker can lead to incorrect or discriminatory choices. In this talk, we will address how artificial intelligence grafts itself into the cycle of decision analysis, presenting stories of both success and criticism, giving particular importance to aspects of transparency, ethics, and attention to the valorization of the human contribution.

This talk explores the evolution of Contract Lifecycle Management toward AI-driven legal solutions, demonstrating how hybrid legal tools can support companies in end-to-end contract management. From document standardization and automated risk analysis to the integration of generative AI and compliance-by-design systems, the session highlights tangible benefits in terms of efficiency, control, and governance of corporate legal processes.

We always talk about A11Y and how to implement usability for users, but almost never do we think about interaction… so why not think of a nice experiment that does it? With "AI on the edge" technologies for facial and eye detection, a bit of linear regression, and a real-time calibration system, I will try—using only JavaScript—to understand the accuracy, current limitations, and potential improvements of an eye-tracking project. Obviously, we hope the demo works…

Text and image generation is entering the age of full virtual-world simulation thanks to Large World Models. In this talk, we will explore how immersive interfaces—powered by models capable of imagining entire virtual contexts—will enable the creation of on-demand virtual spaces. From synthetic training grounds for surgeons and HR professionals to new frontiers in experiential marketing, we will analyze concrete case studies where simulation becomes tangible value. But if reality itself becomes promptable, what are the risks for our identity and our perception of the real world? This session offers a strategic and ethical guide to the next technological convergence.

The demand for AI has exploded, yet the MIT “State of AI in Business 2025” report shows that few projects generate real impact. Companies aren’t looking for prototypes—they want results. This is where Neovia comes in: the value proposition of CRPI Digital Group that turns AI into tangible value. With Ignition, our proprietary framework, we identify the most strategic initiatives; Smart Products provide ready-to-use, customizable, and scalable solutions, while Custom Services support end-to-end adoption. We will present concrete and replicable use cases.

The diverse uses of AI in sports are growing rapidly. This technology—especially through big data analytics—is driving a true revolution, ranging from scouting to performance enhancement, from tactical and strategic decision-making to doping detection (and its misuse). However, to fully leverage the value generated by AI, it is essential to understand its legal implications and the risks in terms of accountability. A challenge that must be addressed in advance.

All companies are talking about artificial intelligence, but few truly manage to transform it into value. The reason is (almost never) technological. In my presentation, I will introduce an AI adoption model based on 10 corporate archetypes: from the Ostrich that avoids the topic, to the Fashionista who chases every novelty, up to the Leapfrog who uses AI to bypass competitors. The archetypes are not just “psychological types,” but a true maturity model: they describe the strategic, cultural, and organizational postures that characterize each horizon of AI adoption. Recognizing one's own archetype means understanding where one stands today and what is needed to move forward. During the speech, I will also offer an overview of the most relevant use cases, divided by corporate functions and sectors, to show how AI generates value when applied to real problems, with significant KPIs and not with vanity metrics.

What do failed robots tell us? Rather than technical limitations, they speak of expectations, cultural imaginaries, and missed connections. This talk explores the failures and promises of social robotics to reflect on how cultural drivers, kindness, and collective choices shape technology. The future is neither unwritten nor predetermined: it is born from and for society, and it can still be steered toward connection, care, and human meaning.

Do you want to better understand how to create the identikit of your buyer personas? Have you ever thought that artificial intelligence can help you find your ideal customer? I will show you step-by-step the research and creation of the Ideal Customer Avatar with AI. Theory, practice, and exercise. We will see examples, templates, and ready-to-use prompts.

95% of AI projects have yet to generate significant returns: the cause is not technological limitations, but conservative architectural choices. The result is a landscape of pilot projects that fail to scale and value that never reaches the customer. This keynote demonstrates how to embed AI into the core of financial services: transforming isolated touchpoints into continuous relationships, periodic checks into real-time supervision, and static rules into intelligent orchestration. Together, Embedded Finance and Generative AI are becoming the new backbone of the financial ecosystem.

AI is not just technology: it’s the heart of transformation. We will show how business challenges give rise to solutions that redefine the future. With concrete examples in the Financial and Industrial sectors, we will demonstrate how AI becomes a competitive advantage and a catalyst for efficiency.

How is a true corporate Artificial Intelligence strategy built? Fastweb+Vodafone shares its journey, founded on sovereign AI and the creation of strategic AI assets to enable secure, responsible, and sustainable innovation. This vision takes shape in the AI applications developed for STEP FuturAbility District, where generative AI is used to create a unique and engaging scientific dissemination experience.

AI is everywhere—but running it efficiently and responsibly in the cloud is a different challenge. As organizations race to deploy AI-powered applications, the energy and resource costs of training and serving models often remain an afterthought. In this talk, we will explore how cloud native technologies can unlock sustainable AI practices that go beyond the hype. We’ll discuss strategies for optimizing AI workloads using Kubernetes, serverless platforms, and cloud native observability, while keeping an eye on carbon impact and resource efficiency. Attendees will learn about scalable MLOps patterns, intelligent workload placement, and green computing techniques that reduce waste without compromising performance. Whether you are a platform engineer, cloud architect, or AI enthusiast, this session will equip you with practical approaches to make AI in the cloud efficient, cost-effective, and environmentally responsible—and finally move the conversation beyond buzzwords.

Mathematical programming methods provide optimal solutions to complex decision-making problems, but they require precise mathematical formulations. As LLMs improve at modeling and translating real-world tasks into these formulations, a new phase of AI emerges: integrating LLMs with mathematical programming tools to create reliable agents capable of making decisions with verifiable guarantees.

The presentation introduces Durana Tech Park, Albania’s first technology and science park, established to position the country as a leading regional hub for technology companies, startups, academic institutions, and digital nomads. It outlines Durana’s strategic vision, governance model as well as the park’s unique fiscal framework with the park expected to host more than 95 resident entities in 2026, further strengthening Albania’s innovation ecosystem and its integration into regional and global technology markets.

This speech will explore the embodiment of AI features in advanced robotics systems and how to design their lifecycle to ensure ethics and safety in the design and adherence to applicable standards and regulation. Also, different AI applications (e.g., preventive safety monitoring, digital twins. neural safety models etc.) to robotics in logistic and industrial applications will be presented to show how they can contribute in improving workplace safety.

In this talk I will show, step by step, what really happens when companies move from theory to practice in complying with AI regulations. Through two real-world cases, we will see where things typically go wrong, how course corrections are made, and what it takes to reach the finish line with method and clarity. A practical journey for those who want to understand how to come out prepared — and alive — on the other side.

Streaming platforms decide what we listen to through opaque, repetitive algorithms. CORRD breaks this pattern: powered by AI, it enables music discovery and personalized recommendations. In this talk, we explore how algorithms shape our choices—and how to give power back to listeners.

If AI becomes the only interface between you and your customers, you’ve already lost.Autonomous agents browse the web, gather data, compare offers, and make decisions on behalf of users — without ever visiting your website.It’s the new “Amazon problem”: disintermediation at industrial scale.Just like Nike left Amazon to regain direct customer relationships, companies today must rethink their websites as agentic systems — not passive showcases, but intelligent architectures able to interact with both humans and AI agents.A strategic framework to help brands retain control over the customer experience in the era of the AI-mediated web.

In this talk, we will explore the foundations of designing reliable and robust Artificial Intelligence (AI) systems. We will delve into key standards and guidelines for AI system evaluation, focusing on technical design and the adoption of practices that integrate trust and reliability from the very early stages of the development process. The session will offer a practical look at the fundamental principles for building AI systems capable of meeting the demands for transparency, accountability, and resilience.

Companies can no longer rely on static documents to assess the reliability of their partners. What’s needed is an infrastructure that certifies data, behaviours, and transactions over time.
This talk explains how KYP integrates AI, blockchain, and verifiable credentials to create a dynamic and immutable registry of supplier compliance, enabling more ethical, secure, and fully verifiable collaborations.

What does it mean to be a legal professional in the era of AI? Which activities can be enhanced and which skills are emerging? Paolo Fois explores the transformation of the sector, comparing the types of legal AI available and showing how it can reduce time, simplify processes, and increase the productivity of daily legal work.

Niuexa presents AEO Analyzer, an innovative SaaS platform designed to optimize website visibility in next-generation AI search engines. By analyzing AI signals, semantic quality, visibility, and content, AEO Analyzer helps businesses improve how artificial intelligences understand and trust their websites. Get detailed reports and strategies to make your site “AI-ready” and competitive in the evolving digital landscape.

In a world saturated with data, AI models, and complexity, the challenge is not adding technology but distinguishing the signals that make a difference from the noise. Starting from the physics of complex systems, an approach emerges where noise becomes a resource to be governed to support real decisions. Humans.tech will bring to the stage what becomes possible through the development of intelligent agents: making useful signals readable. Decoding before coding.

The greatest achievement in sports? Creating environments where children feel safe and respected. This means learning how to address potential issues that may arise within the sports context, such as bullying, but also being able to recognize what is happening in other areas of a young person’s life. In this panel, we will explore how everyone can play their part by combining experience, ethics, and technology. Because AI, too, can become our ally—if we train it well.

Generative AI has zeroed out the marginal cost of creativity. Analysis and generation, once distinct skill sets, are now available "on tap." However, the true revolution doesn't lie in orchestrating AI agents faster — it lies in shifting from selecting content to generating hyper-personalized experiences. It’s no longer a matter of "which of these 15,000 pieces of content should I show you," but rather "what content should I create for you." Content, software, and interfaces generated in real-time, for that specific person, at that exact moment. Hyper-personalization is no longer a goal to be chased — it is the natural consequence of Just-in-Time creativity. Federico Perazzi, with 15 years of AI experience across Disney Research, Adobe, Meta, and Bending Spoons, explores how this shift will transform products, experiences, and the very role of human beings.

We built a multi-vertical SaaS platform in 8 months (60% faster than the average). Evolving from digital memorials to a "Life OS" (DNA, wellness, insurtech), our team of 15 people orchestrates 40+ AI agents. We will explore the "Human-First" model, where AI acts as a collaborator. We will analyze the data flows that transform AI from a feature to a secure execution layer.

AI is leading coding into a new era, opening up programming to a much broader and less experienced audience. Join us to discover how to use an IDE and an AI agent to produce high-quality analysis and output using only natural language.

CLARA is a real-time AI-assisted cytology platform designed to bring high-quality medical screening closer to patients. This talk explores how robotics, telemicroscopy, and artificial intelligence can democratize access to fast, accurate diagnostics, reduce inequality in healthcare, and transform population-level screening into an accessible, point-of-care reality.

New AI-based language technologies, such as ChatGPT, are transforming our everyday lives: they inform, advise, and influence values and decisions. Drawing on recent research findings, we will explore how people actually use these systems, what risks emerge in everyday interactions, and what new directions are opening up. An invitation to look beyond the technology toward its real effects on society.

A hands-on workshop dedicated to professionals, managers, and founders involved in commercial activities. This session is designed to demonstrate how AI is reshaping the entire sales cycle. Participants will gain a clear and practical roadmap of exactly where and how AI supports sales efforts: from analyzing market trends and the competitive landscape to preparing for high-stakes client interactions and positioning products or services with true relevance.We will dive deep into one of the most critical stages of the sales process: transforming raw data— such as conversation notes, documents, and market signals—into a coherent, customer-centric commercial message. This is the core focus of our practical exercise. The goal is to provide participants with a practical, replicable method to use AI as a partner in strategic reasoning and commercial positioning, ensuring your outreach never feels generic or standardized. This workshop is reserved for holders of an AI Festival 2026 ticket and also requires a dedicated ticket, available for separate purchase, to access the individual session.

When we talk about "agentic AI," we usually assume that agency remains a human property—extended, amplified, or delegated to increasingly autonomous systems. But what happens when action no longer emerges from a single deciding subject, but from a network of dialogues between multiple subjective states and non-human intelligences? The Jianwei Xun experiment—a book on reality manipulation written with artificial intelligences and attributed to a fictional author built through the same mechanisms it analyzed—demonstrated how authorship can become unassignable, distributed across humans, systems, and training corpora. This is not a matter of hybrid teams where humans and algorithms collaborate while maintaining distinct roles: it is the emergence of a Beyond-Subject, a conversational entity that belongs neither to one nor the other, but is constituted in between, in the space of encounter. The real challenge is learning to consciously navigate this condition of permanent co-production, developing a “double play of consciousness”: the ability to be simultaneously inside and outside the dance with machines.

In 2026, the AI landscape is shifting from "experimentation" to "industrial governance." However, the ground beneath us is far from stable. With a hardware oligopoly, a fragmented software market, and a relentless cycle of research-driven disruption, leadership has become a high-altitude challenge. How do you steer a global organization when the wind is at its strongest? Join us as we move beyond the hype of speed and focus on the necessity of direction, providing the tools to navigate the most turbulent transition in tech history.

AI capabilities are advancing rapidly, but history shows that transformative technologies don’t deliver productivity gains automatically. The real challenge lies in how organizations adapt. In this session, Nestor Maslej, the former editor-in-chief of the Stanford AI Index and contributor to the forthcoming International AI Safety Report presents a data-driven view of where AI stands today and what comes next. The talk examines four forces shaping the AI era: accelerating technical performance, intensifying U.S.–China competition, a 2025 inflection point in business adoption, and the growing gap between AI availability and effective integration. Attendees will leave with a clear framework for the questions business leaders must answer now: from productivity and labor impacts to build-versus-buy decisions and responsible deployment.

We are entering a new era of search capabilities that were unimaginable just five years ago. As user behaviors shift from keywords to conversations, tech giants are locked in a high-stakes race for dominance. This session explores the evolution of human intent and provides actionable insights for brands: how can businesses maintain visibility and relevance in an AI-driven "Answer Engine" landscape?