# Exploring AI in Entrepreneurship, Innovation, and Leadership in the Digital Era  I had the incredible opportunity to participate in the **DigiVisions 2024** event, focusing on entrepreneurship, innovation, and leadership amidst the digital transformation, organized by UNIL and EPFL. During this enlightening conference, I gave an interview alongside Laurent Haug, moderated by Quentin Gallea. It was a privilege to be part of such a dynamic discussion, featuring numerous fascinating speakers and a variety of stimulating topics. <iframe width="100%" height="530" src="https://www.youtube.com/embed/y9aOFewqdxs?t=3842" title="DIGIVISIONS 2024 - Entrepreneuriat, innovation et leadership à l&#39;ère de la transformation numérique" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> ## Key Insights from the Conference ### AI's Role in My Research At the conference, I introduced my work as a researcher, where **half of my efforts involve developing algorithms, creating research, coding to train models, evaluating them, and examining their impact on the world and humans**. The other half focuses on translating these findings into articles that are both clear and scientifically rigorous, thereby conveying my conclusions and results effectively. Interestingly, **both aspects of my work are being transformed by AI**. For instance, in coding, I now use a tool called Cursor, powered by AI. This has dramatically altered my workflow; I interact with AI to generate code rather than writing it manually. While this raises questions about the pros and cons, it's undeniable that a significant portion of my work has been replaced.  ### AI-Generated Research and Its Implications Similarly, the second part of my work—writing scientific articles—is also evolving. Recently, an article discussed a multi-agent AI system capable of conducting research from start to finish. This includes initiating research topics, conducting experiments, analyzing results, and drafting the article. Although currently, these AI-generated articles may lack depth, they demonstrate the growing capability of AI in academic research. <iframe width="100%" height="530" src="https://www.youtube.com/embed/iC-wRBsAhEs" title="OpenAI&#39;s ChatGPT Does Research… And Breaks Itself!" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe> ### Is My Work Disappearing? Is it problematic that my work could be at risk of disappearing? In reality, it's not a major concern for me personally. We are experiencing a transformative period where AI acts as a **supportive assistant** rather than a replacement. We are in a phase of *co-piloting*, where AI enhances and multiplies our productivity. In my everyday experience, the productivity boost I achieve through AI is substantial. The demand for AI researchers is increasing, not decreasing, even as AI takes over some tasks. This paradox highlights that, despite AI undertaking part of their work, the role of AI professionals is more vital than ever. Thus, we are in a transitional phase where AI does not replace human work but supports and amplifies it.  ### Understanding the Future of AI: Levels of Development To understand where we stand in AI development, it's instructive to consider OpenAI's framework toward Artificial General Intelligence (AGI). They outline five levels of AI capability: 1. **Level 1: Chatbots** Systems that mimic human conversation and pass the Turing test. ChatGPT is a prime example. 2. **Level 2: Reasoners** Models like GPT-4o perform complex reasoning and provide thoughtful responses to complicated questions. 3. **Level 3: Agents** These AIs take actions and learn from their outcomes, such as OpenDevine, which can autonomously develop a project like a computer game by iterating coding, testing, and refining. 4. **Level 4: Innovators** AIs capable of creating entirely new inventions across multiple domains, exemplified by systems like AlphaFold in biomedical research. 5. **Level 5: Organizations** Envisioned as the AI pinnacle, where AI manages entire operations, achieving transformative impacts on AI capability. Despite rapid progress, achieving AGI remains quite distant. We're currently transitioning from level 2 to level 3, indicating both remarkable accomplishments and significant challenges ahead. <a href="https://openai.com/index/planning-for-agi-and-beyond/" target="_blank" style="text-decoration: none; color: inherit;"> <div class="card" style="display: flex; align-items: center; padding: 10px; background: #fff; border-radius: 8px; box-shadow: 0 4px 8px rgba(0,0,0,0.1); max-width: 400px; margin-bottom: 10px;"> <div class="icon" style="font-size: 24px; margin-right: 15px; color: #0056b3;">🧠</div> <div class="card-content" style="flex-grow: 1;"> <div class="card-title" style="font-size: 18px; font-weight: bold; color: #333;"> Planning for AGI and beyond | OpenAI </div> </div> </div> </a> <a href="https://news.mit.edu/2024/president-sally-kornbluth-openai-ceo-sam-altman-discuss-future-ai-0506" target="_blank" style="text-decoration: none; color: inherit;"> <div class="card" style="display: flex; align-items: center; padding: 10px; background: #fff; border-radius: 8px; box-shadow: 0 4px 8px rgba(0,0,0,0.1); max-width: 400px;"> <div class="icon" style="font-size: 24px; margin-right: 15px; color: #0056b3;">🤖</div> <div class="card-content" style="flex-grow: 1;"> <div class="card-title" style="font-size: 18px; font-weight: bold; color: #333;"> President Sally Kornbluth and OpenAI CEO Sam Altman discuss the future of AI | MIT News </div> </div> </div> </a> ## Addressing Key Questions Throughout the conference, I was asked two pivotal questions: 1. **Will AI replace us?** 2. **In education, which parts can be replaced by AI, and which should remain under human oversight?** In this article, I will primarily focus on the first question, as it necessitates extensive argumentation and synthesis of ideas we've explored. The notion of AI replacing human roles is profound and complex, demanding thorough exploration to understand the implications and the transitional phase we currently navigate. ### The Human Role in an AI-Driven Future Although AI currently acts as an aide—helping with tasks like drafting emails, coding, and brainstorming—**humans remain indispensable in the loop**. The incredible efficiency we achieve today is a result of human-AI collaboration. Attempts to fully automate jobs with AI often fall short; the real value lies in enhancing human capabilities through AI support. In this co-piloting model, AI handles repetitive, uninteresting tasks, leaving humans to focus on creativity and decision-making, such as crafting the response rather than worrying about typing numerous emails. The centrality of human oversight is crucial to achieving incredible effectiveness. ### Imagining a Future with General AI The hypothetical arrival of AGI raises concerns about job displacement and societal restructuring. However, it's crucial to recognize that our current social constructs, such as employment, are based on value production driven by survival needs, like food and shelter. In a world where AI independently generates value, the concept of employment could fundamentally change. We might witness an immense societal transformation where the focus shifts from individual value creation to perhaps leisure or innovation without the pressure of traditional work.  ### Finding Enjoyment Beyond Value Creation Beyond the need to create value, activities that don't traditionally generate value hold immense interest. Take chess, for instance. Despite AI systems like Deep Blue mastering the game—or appearing unbeatable—people continue to play. The real thrill lies not in defeating a machine, but in the joy and intellectual challenge chess provides. Similarly, **AI has democratized learning**, granting anyone access to high-level training resembling 'a grandmaster at home.' This accessibility has led to a surge in chess players benefitting from insights and feedback previously unavailable. This phenomenon reflects a broader trend where AI levels the playing field, particularly in education, by enabling everyone to achieve similar proficiency levels. This transformation allows for the enjoyment and progression of skills without direct economic value. Moreover, artists continue creating despite AI producing extraordinary works. Many find joy in crafting art purely for personal satisfaction. **AI assists and inspires rather than replaces them**. These pursuits remain valuable for personal fulfillment, even if not economically vital.  ### The Significance of Human Imperfection A crucial element of appreciation in art and activities is human imperfection. The allure of artistic content, threatened by AI, is rooted in its imperfections. Humans still prefer playing chess against other humans or more 'human-like' AI opponents because their mistakes and unique play styles offer a richer experience. In the realm of art, this appreciation for imperfection persists. Consider photorealistic paintings—they captivate us not just because they mimic photographs, but because each brushstroke reflects human effort and individuality. This emphasis on imperfection underscores a broader truth: **what truly differentiates human creations is their unique, imperfect nature**. As AI becomes more ubiquitous, whether it's composing emails or crafting art, what will continue to set human work apart is this personal touch. It's the ideas, choices, and nuances humans bring to their work that maintains our connection to it, even when AI handles many foundational tasks.  ### The Challenge of Rapid Technological Change The real challenge with AI is not its emergence but the **pace of change** it introduces. Historically, new technologies like the Internet have brought immense transformation, yet society adapted over time. The issue lies in the speed at which these changes occur. If AI development proceeds too quickly, especially with the potential advent of AGI, the societal shift could be abrupt and disruptive. A slow and steady approach allows society to adjust and integrate these advancements without upheaval. While I am not overly concerned about the eventual transformation, the transition period is what warrants consideration.  --- This experience highlighted the necessity of thoughtfully integrating AI into our lives, especially in educational and entrepreneurial contexts, to harness its full potential while addressing societal needs. By embracing both the capabilities of AI and the unique imperfections of human effort, and monitoring the pace of technological advancement, we can find balance that fosters growth and stability.