Swetlana AI Podcast

In this episode, we explore the New York Times article that highlights the growing concern over AI energy consumption, particularly with the rise of Nvidia's energy-intensive chips. As AI systems continue to expand, the AI carbon footprint grows, raising questions about the environmental impact of this technology. The discussion also delves into how much energy AI consumes, especially when it come...

In this episode, we dive deep into the life and career of Sam Altman, a key figure in the world of technology and artificial intelligence. Drawing on two insightful sources—an in-depth conversation between Altman and Debora Spar at Harvard Business School and a detailed biography from Wikipedia—we trace Altman’s journey from a computer-obsessed teenager to his current role as the CEO of OpenAI. Th...

I'm delighted to announce that NotebookLM can create explicit content now! I'm seeing unlimited possibilities. Enjoy this little piece of "critique" of Elon Musk's Twitter. Hosted on Acast. See acast.com/privacy for more information.

So the other day Apple had published a paper called "GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models" (by Iman Mirzadeh, Keivan Alizadeh, Hooman Shahrokhi, Oncel Tuzel, Samy Bengio, Mehrdad Farajtabar). That's what we're looking at in this podcast episode. This research paper investigates the reasoning capabilities of large language models (LLMs) by e...

In this episode, we dive into the enduring mystery surrounding the identity of Satoshi Nakamoto, the enigmatic creator of Bitcoin. We explore key insights from Satoshi’s early communications with key figures in Bitcoin’s development, offering a rare glimpse into the mind behind the cryptocurrency revolution. We also discuss the highly anticipated HBO documentary, Money Electric: The Bitcoin Myster...

I told NotebookLM to make a podcast about itself. Here's the result. The hosts appear (only appear, obviously) to be self-aware. *** The hosts of this discussion are AI entities created by Google's NotebookLM. They are not human, but are designed to discuss and analyse information. This discussion is an example of NotebookLM's Audio Overviews feature, which can generate AI podcasts from uploaded d...

In this episode of Swetlana AI Podcast, we explore the fascinating world of tokenization in AI, a process fundamental to how large language models (LLMs) like those from OpenAI function. What are tokens, and why do they matter in AI? We’ll break down the basics, from defining tokens to discussing how text is transformed into units that an AI model can process. Whether you're familiar with AI or ju...

In this episode we're looking at Steve Jobs’ keynote address at MacWorld 2007, in which he unveiled the first iPhone. The presentation of iPhone 1 focuses on the phone's innovative features, including its revolutionary multi-touch interface, intuitive user experience, and integration with Apple's software and services. Jobs also highlights partnerships with major companies like Google, Yahoo, and...

Today we’re discussing Bill Gates’ 1994 speech, “Information at your fingertips”: https://www.youtube.com/watch?v=7fJWMsgxzvA The speech was delivered at the Comdex trade show. Gates describes his vision for the "information highway," a future where computers and networks will enable pervasive communication and information access. He paints a picture of a world where people use mobile displays, co...

In this episode, we explore the Content Authenticity Initiative (CAI) and the Coalition for Content Provenance and Authenticity (C2PA), a collaboration aimed at combating disinformation through a provenance metadata standard. Learn how this technology allows users to trace the origins and edits of digital content—photos, videos, and audio files—using metadata secured with hashcodes and digital sig...

In today's episode, we dive into the essential skill of  prompt engineering —a key practice for effectively working with large language models (LLMs). We'll explore strategies and techniques for crafting prompts that lead to high-quality AI-generated responses. From using context and examples to advanced techniques like chain-of-thought prompting and fine-tuning, we break down the best practi...

Today we’re discussing Dead Internet Theory , a conspiracy theory which posits that the majority of online activity is generated by automated bots rather than real people.  This theory asserts that these bots are intentionally deployed by various actors, including governments and corporations, to manipulate users and spread misinformation.  The texts explore evidence supporting the theor...

Sam Altman, the founder of OpenAI author of "Moore’s Law for Everything" ( https://moores.samaltman.com/ ), proposes a plan to use the wealth generated by artificial intelligence to benefit everyone. He argues that, as AI automates many jobs, the price of goods and services will plummet, drastically increasing societal wealth. To ensure this wealth is shared, he advocates for a new social contract...

In this episode, we dive into two fascinating case studies on the use of AI in the legal industry. We're looking at a comprehensive guide for legal professionals on harnessing the power of ChatGPT for tasks like case law analysis, statutory interpretation, and legislative history research. We’ll also examine the ethical concerns surrounding AI in law—privacy issues, biases, and the need for transp...

In this episode, we explore the transformative potential of artificial intelligence (AI) on global economies and the future of work. From boosting productivity to creating new jobs, AI promises economic growth, yet it also presents challenges, such as job displacement and ethical concerns. We’ll break down how AI could contribute trillions to the global economy by 2030, impact job markets across a...

In this episode we're exploring the growing use of AI in music generation, highlighting its potential and limitations. Here are the sources we're discussing: One source explains the technical details of diffusion models used in platforms like MusicLM and Stable Audio, showcasing how they learn to generate music based on textual prompts and melodic input. Another article discusses the Grammys' stan...

In this episode, we dive into the escalating conflict between traditional artists and the rapidly advancing world of AI-generated art. We'll explore key developments, including the rise of the "Artists Against AI" Facebook group, which advocates for preserving human creativity in the face of AI's expansion. We’ll also discuss OpenAI’s latest creation, DALL-E 3, and its developers' efforts to addre...

“Attention Is All You Need” (by Vaswani et al.) is an academic paper published at the 2017 Neural Information Processing Systems (NIPS) conference. It's one of the most important papers on the topic of AI, because it has introduced a groundbreaking architecture in natural language processing (NLP) and machine learning. In this episode we are discussing the key points of the paper. The key idea of...

Today we're discussing the concept of "founder mode," a management style where a company's founder maintains a hands-on approach at all levels, often involving direct oversight of teams and decision-making. This style is contrasted with "manager mode," which prioritises delegation and empowers teams to take ownership of their work. While some argue that "founder mode" fosters innovation and succes...

Welcome to Episode 2 of my Swetlana AI podcast. This podcast explores the evolving landscape of Artificial Intelligence (AI) in the German engineering sector, focusing specifically on the perspectives and experiences of engineers. Key questions addressed in the podcast include: ● How do engineers in various roles and industries define AI? ● What are their experiences with existing AI applications?...

Excited to share my first podcast episode - it's about AI art. In this episode we will be talking about AI art and its legitimacy, and we'll look at questions such as: - Can AI make art? - Is AI art = theft? - Is AI art copyrighted? - Is AI art bad? - Will AI make artists obsolete? Papers that are helpful for this discussion: - Elzė Sigutė Mikalonytė/Markus Kneer, "Exploring the Role of Intention...