Explorez tous les épisodes du podcast AI for All

This 2nd episode on a 2 part series analyzes how GenAI has evolved to deliver automation through AI Agents and Agentic AI. It highlights the differences between both of them citing many sources from IBM, Anthropic McKinsey and Gartner amongst others as well as their best use cases and differences and also mentioning some of the risks.

1. Generative AI (GenAI)

Generative AI refers to reactive systems that wait for user input, typically a prompt, to generate content1. Their job is to generate some kind of content based on what was provided in the prompt.

2. AI Agents

AI Agents are typically built to do specific tasks5. They are designed to help users with actions like answering questions, organizing a calendar, or managing an email inbox. Think of them as virtual helpers that do exactly what you tell them to do, without thinking for themselves. They are great at automating simple, repetitive tasks but don't have the autonomy or decision-making abilities that Agentic AI does

3. Agentic AI

Agentic AI is a type of AI that is all about autonomy16. It refers to AI systems that possess the capacity to make autonomous decisions and take actions to achieve specific goals with limited or no direct human intervention17.... The term "agentic" refers to these models' agency, or, their capacity to act independently and purposefully18.

Unlike AI Agents built for specific tasks, Agentic AI can make decisions, take actions, and even learn on its own to achieve specific goals16. They are proactive systems,

This first episode on a 2 part series analyzes how GenAI has evolved to deliver automation through AI Agents and Agentic AI. It highlights the differences between both of them citing many sources from IBM, Anthropic McKinsey and Gartner amongst others as well as their best use cases and differences and also mentioning some of the risks.

1. Generative AI (GenAI)

Generative AI refers to reactive systems that wait for user input, typically a prompt, to generate content1. Their job is to generate some kind of content based on what was provided in the prompt.

2. AI Agents

AI Agents are typically built to do specific tasks5. They are designed to help users with actions like answering questions, organizing a calendar, or managing an email inbox. Think of them as virtual helpers that do exactly what you tell them to do, without thinking for themselves. They are great at automating simple, repetitive tasks but don't have the autonomy or decision-making abilities that Agentic AI does

3. Agentic AI

Agentic AI is a type of AI that is all about autonomy16. It refers to AI systems that possess the capacity to make autonomous decisions and take actions to achieve specific goals with limited or no direct human intervention17.... The term "agentic" refers to these models' agency, or, their capacity to act independently and purposefully18.

Unlike AI Agents built for specific tasks, Agentic AI can make decisions, take actions, and even learn on its own to achieve specific goals16. They are proactive systems,

Quantum computing is rapidly evolving from a theoretical concept to a tangible technology with the potential to revolutionize various fields, especially AI. Significant progress has been made in qubit technology, error correction, and algorithm development. This technology is poised to accelerate AI advancements in areas like machine learning, optimization, cryptography, and drug discovery.

Key Themes and Ideas:

1. Challenges and Solutions:

• Error Sensitivity: Quantum computers are highly susceptible to noise and environmental disturbances, leading to computational errors. "One of the biggest challenges with quantum computers is that they’re incredibly sensitive to the smallest changes, or ‘noise’ in their environment."
• Qubit Coherence and Stability: Maintaining qubit coherence for sufficient computation time is a significant hurdle.
• Scalability: Building larger, more stable quantum systems while reducing error rates remains a challenge.

1. Impact on Artificial Intelligence:

• Accelerated AI Advancements: Quantum computing has the potential to drastically accelerate AI capabilities in various areas.
• Machine Learning: Quantum machine learning models can improve natural language processing, image recognition, and autonomous systems.
• Optimization: Quantum optimization algorithms can enhance supply chain management, logistics, and financial modeling.
• Drug Discovery and Materials Science: Quantum simulations can accelerate drug discovery, materials science, and climate modeling.
• Cryptography: Quantum cryptography offers robust security solutions for data protection and communication.
• Uncertain Timeline: The exact timeline for a quantum "singularity" (a point where quantum computers achieve practical, widespread utility) is uncertain and debated.

Conclusion:

Quantum computing is a dynamic and rapidly advancing field with significant potential to transform AI and various other industries. Overcoming current challenges related to qubit stability, error correction, and scalability will be crucial for realizing this potential. The substantial investments and ongoing research efforts by leading technology companies and governments suggest that practical, fault-tolerant quantum computers are becoming increasingly likely in the coming years.

DeepSeek, a Chinese AI startup, is rapidly gaining prominence by developing and openly releasing highly competitive, cost-effective AI models. Its innovative techniques, including reinforcement learning and a mixture-of-experts architecture, allow for superior performance while minimizing training costs. This has disrupted the AI market, sparking price wars and forcing established companies like OpenAI and Google to adapt. While facing challenges such as access to advanced computing resources and potential censorship concerns, DeepSeek's open-source approach and focus on efficiency are reshaping the AI landscape. The company's strategic partnerships and rapid innovation position it as a major player to watch.

This episode explores not only the overall AI Landscape advancements of 2025 but it builds upon the AI evolution in 2024 and its incredible advancements.

Some key likely trends that we will see AI evolving into during 2025 are:

1. Clearly its pervasiveness and growing ubiquity in everyday life.
2. The growth of Generative AI and its increased capabilities
3. AI influenced if not determined healthcare
4. Continued AI Relevance and adoption in Education
5. And also AI Driven Environment Relevant systems - key word "sustainability"

... Maye even Quantum AI?...

To learn more listen to this unique view of the likely trends we will experience with the Advancement of AI in 2025.

And with that, thank you for your audience in 2024 and wish you a healthy and successful NEW YEAR 2025! May the worth of AI and all its benefits be with you!!

This Episode summarizes the main milestones that AI achieved during the year 2024. From becoming a promising advancement to receiving 2 Nobel Prizes, one for Physics and one for Chemistry. Of course too many other advancements in this field have taken place in 2024, nevertheless in this case we review some of the major events including the new LLM by Google Gemini 2.0 Trillium Chip making an incredible accelerator in AI Model Performance, The Nobel Price in AI Ethics, a must that AI should never be without in its Governance Framework, The Centerstage take over by Gen AI as well as medical applications. The investment focus and competition that AI has driven as it establishes itself as a mainstream technology with true scientific and practical applications across the board. And of course also making reference to real challenges, be it in bias or displacements in the job market, concerns about regulatory, ethics and privacy concerns.

The podcast explores the different types of artificial intelligence (AI) and their capabilities.

It distinguishes between narrow AI (weak AI), which performs specific tasks, general AI (strong AI) with broad human-like cognitive abilities, and superintelligent AI (ASI), which surpasses human intelligence in all aspects.

It also explores the AI categories based on functionality (e.g., reactive machines, limited memory, theory of mind), learning models (supervised, unsupervised, reinforcement, and deep learning),

It exemplifies specific branches of AI like machine learning, robotics, and natural language processing.

Finally, it concludes with an examination of Artificial General Intelligence (AGI), exploring its potential benefits, risks, and societal implications.

For the First Time in History the Nobel Prize in Physics is awarded for work related to AI.

This now gives AI and its related fields of study a new place in Academic History

The 2024 Nobel Prize in Physics was awarded to John Hopfield and Geoffrey Hinton for their groundbreaking work in artificial neural networks. Hopfield, known for his Hopfield network, focused on bridging the gap between physics, biology, and computer science, using insights from the brain to improve AI. Hinton, considered the "godfather of AI," developed methods allowing computers to recognize patterns in data, revolutionizing fields like speech and image processing.

Both researchers, while optimistic about AI's potential, have raised concerns about the technology's rapid advancement, Inevitably as it may be in its benefits to humanity, the further this technology goes the more its potential to transform society as a whole. Present concerns range from its potential for job displacement, misuse, and the possibility of AI exceeding human intelligence. At the same time its benefits are far outreaching and will continue to demonstrate its incredible capacity for scientific and technological breakthroughs.

The Evolution of Artificial Intelligence: From Philosophical Roots to a Transformative Future

This podcast explores the historical development, key figures, and potential future of artificial intelligence (AI),

The Essence of AI:

AI empowers computers to mimic human intelligence, tackling tasks like problem-solving, learning, and language comprehension. These systems analyze data to make informed decisions, often exceeding human capabilities in specific areas.

Ancient Foundations and Philosophical Ponderings:

The roots of AI stretch back to antiquity, with Greek philosophers like Aristotle laying the groundwork for reasoning and logic. This foundation was further built upon in the 16th century by:

• René Descartes, who investigated the mind and the possibility of machines replicating human thought.
• Gottfried Wilhelm Leibniz, whose work on mathematical logic and mechanical calculators foreshadowed computational advancements.

Emergence of the "Artificial Brain":

The early 20th century witnessed the conceptualization of artificial humans, prompting inquiries into the feasibility of creating "artificial brains". Notably:

• Charles Babbage developed the first programmable computer, paving the way for digital computation.
• Ada Lovelace, recognizing the potential of Babbage's invention, theorized generalized computing machines capable of tasks beyond mathematical calculations.

Birth of AI as a Discipline:

The mid-20th century marked the formal establishment of AI as a distinct field of study. Key contributions include:

• Alan Turing introduced the Turing Machine, a theoretical model for computation, and the Turing Test, which assesses a machine's ability to exhibit human-like intelligence.
• John McCarthy, widely considered the "father of AI," coined the term "artificial intelligence" and organized the 1956 Dartmouth Conference, a pivotal event in AI history.

McCarthy's Legacy:

John McCarthy's impact extends beyond terminology. His development of the LISP programming language, specifically designed for AI research, remains significant. Additionally, his concept of time-sharing revolutionized interactive computing.

Timeline of Progress:

The evolution of AI can be traced through distinct periods:

• 1950s: Following the foundational work of pioneers like Čapek and McCarthy, symbolic reasoning emerged.

AI Today and Tomorrow:

AI is now ubiquitous, powering virtual assistants, e-commerce search engines, and even self-driving cars. Its applications span diverse industries, from healthcare to finance. However, ethical considerations surrounding fairness, transparency, and accountability are paramount.

This episode explores the milestone of GenAI passing as a human by completing the Turing test as well as why this milestone is so relevant, It compares the 2 giants ChatGPT and Gemini in this arena whilst at the same time explores the ethical dilemma of what it would mean if we weren't able to distinguish AI from humans and how this is a risk when misused.

The source, written by Esteban Remecz, a CIO VP at Maxion Wheels, explores the likely trends of Artificial Intelligence (AI) in 2024 and beyond. Remecz emphasizes the exponential growth of AI across various industries, predicting a possible singularity in 2026-2028.

The author discusses the potential benefits of AI in fields like healthcare, education, and cybersecurity, while also addressing the ethical and legal challenges associated with its increasing influence.

He highlights the importance of developing ethical frameworks and regulations to guide the responsible development and application of AI.

Finally, Remecz identifies 15 key areas where AI will dominate in 2024, including AI-driven automation, creativity, and cybersecurity.

1. AI Regulation in Life Sciences: The EU AI Act is a key regulatory development focusing on a risk-based approach to AI governance, particularly in high-risk areas like healthcare. Switzerland, a prominent AI hub, faces the challenge of balancing innovation with international regulatory alignment.
2. Managing AI Risks: The EY Trusted AI Framework helps organizations understand and mitigate AI risks across various dimensions, including performance, bias, resilience, explainability, and transparency. Integrating AI risk management with existing ERM programs is crucial.
3. Black Box Models & Challenges: While AI offers significant potential for life sciences, "black box" models raise concerns about interpretability, potential bias, data privacy, and ethical considerations, necessitating careful mitigation strategies.
4. Building Trustworthy AI: Life science companies must prioritize building trust and transparency into their AI systems. This involves addressing potential issues related to human oversight, data quality, modeling techniques, and comprehensive documentation. Collaboration with experts can help navigate the evolving regulatory landscape.