Co-creating with AI

In our latest "Co-Creating with AI" episode, we explore the nuances of AI models in coding. We differentiate between training a model from scratch and fine-tuning an existing one, emphasizing that fine-tuning builds on a model's current capabilities. Our discussion highlights the increasing role of AI in mainstream industries, exemplified by Coca-Cola's new AI co-created flavor. Throughout, we exchange insights and delve into the growing influence of AI in co-creation, showcasing the evolving synergy between humans and AI.

Hosted on Acast. See acast.com/privacy for more information.

In today's episode, we delve deep into the transformative impact of AI on the job market, particularly in the realm of knowledge work. We discuss how AI is not just a tool for specialists but also a game-changer for generalists, enabling them to become more specialized in various fields without spending years acquiring expertise. We're fascinated by the idea that AI can make us "generalist specialists," allowing us to be competent in multiple areas at once.

We also explore the adoption rate of AI tools like ChatGPT, noting that while it may seem like "everyone is using AI," the reality is that we're still in the early stages of the adoption curve. According to a report, only 2% of the U.S. population has tried ChatGPT, and 15% of those found it to be a game-changer. We ponder whether this slow adoption rate is due to a lack of awareness or perhaps even fear of AI, perpetuated by "AI doomers" who may want to keep the technology's advantages to themselves.

We're particularly intrigued by the global implications of AI adoption. We discuss how AI could serve as an equalizer in the global job market, much like globalization did for manufacturing. For instance, a copywriter in India could potentially produce work at a quality and speed that would make them competitive with copywriters in more developed markets, thereby increasing their earning potential.

We also touch on the future, speculating on how growing up as "AI natives" might affect the next generation's approach to work, education, and life in general. We consider the rapid pace of technological change and how it's crucial for educational institutions to keep up.

In summary, we believe that AI has the potential to be a great equalizer in the job market, but also acknowledge that we're still in the early stages of this transformation. Whether it's a boon or a bane will largely depend on how quickly people adapt to it and how equitably its benefits are distributed. It's a topic that raises more questions than answers, but one thing is clear: the impact of AI on the job market is a subject that warrants ongoing discussion and scrutiny.

Hosted on Acast. See acast.com/privacy for more information.

In this episode of our podcast, we engage in an in-depth exploration of how AI models, particularly language models like GPT-3 and GPT-4, understand and handle context. Our conversation unravels the complexities of these models, the data they're trained on, and how they use this data to generate responses. We make it clear that these models don't understand context the same way humans do, lacking an innate understanding of the world, personal experiences or an internal model of reality.

We delve into the reality of AI, dispelling the notion that these models 'think' in the way humans do. We emphasize that AI models are impressive pattern matchers but are essentially devoid of any genuine understanding or consciousness. They are simply algorithms running on massive amounts of data and don't possess beliefs, feelings, or intentions, despite sometimes appearing to.

We discuss the fascinating aspect of co-creation, in which users and AI collaborate, using the model's capabilities to create new content, design systems, write code, and more. As we engage with the AI, our input provides the context necessary for the AI to generate relevant and meaningful output.

Moreover, we stress the importance of transparency in AI's thought processes. As AI models are used for more advanced tasks, the ability for them to explain their decisions becomes more vital. However, we acknowledge the challenge that the reasons AI generates may not necessarily align with its actual decision-making process, as understood by humans.

Lastly, we highlight the power of the chat interface as a medium for AI interaction. We agree it's like the first step, akin to the terminal in the early internet days, and anticipate a future where more advanced, co-creative interactions are commonplace.

We conclude the episode by emphasizing the beautiful complexity of developing AI services and the exciting challenge of designing user experiences that take both technical aspects of AI and human behavior into account. The next big user experience for AI remains an open, tantalizing question.

Hosted on Acast. See acast.com/privacy for more information.

In this episode of "Co-Creating with AI," we have our guest Malcolm Sparks, the founder and CTO of JUXT, discussing the database they use called xtdb. The conversation revolves around the early days of the internet and the concept of co-creation. We explore how the web initially started as a read-write platform, where websites were like Google Docs that anyone could edit. It was a more open and collaborative environment compared to the broadcast-style mass media. We touch upon the transition to web 2.0, where social media and platforms allowed more people to become publishers, but the true essence of co-creation started to fade away.

Malcolm shares his personal experience of discovering the web in the mid-90s, where Netscape Navigator and Mosaic were popular browsers. He describes the excitement of exploring new websites every day and the freedom to edit HTML pages as if they were Google Docs. The web was all about linking and discovering new content, promoting an open and surf-like experience. However, as the web evolved, the focus shifted towards complex web apps with limited editing capabilities, making it challenging to learn from the source code.

The conversation then moves towards xtdb, a database developed by JUXT, which combines the best features of relational and graph databases. Malcolm explains that xtdb embraces the free-form and schemaless nature of the early web, allowing users to store documents of any structure. It provides the flexibility to evolve and change data models over time, just like how websites were constantly updated and linked in the early days. Xtdb enables co-creation by empowering users to adapt and refine their understanding, make mistakes, and change their minds without being bound by rigid structures.

The episode delves into the unique aspects of xtdb, such as its graph database functionality and bidirectional links. We discuss the significance of bidirectionality in knowledge management and the challenges of implementing it in an open system like the web. Malcolm also touches upon the concept of property graphs, where every value in a document can optionally serve as the identity of another document, allowing for flexible and interconnected data models.

The episode highlights the co-creative nature of the early web and how it influenced the development of xtdb. The discussion revolves around the importance of open and collaborative environments, the evolution of data models, and the need for flexible databases that support co-creation and adaptability. We want to thank our guest, Malcolm Sparks, for his insightful contributions and also to our listeners for their continuous engagement.

Hosted on Acast. See acast.com/privacy for more information.

In this thought-provoking episode, we delve into the innovative intersection of AI and creativity. From discussing AI as a co-creator rather than a tool, to exploring the nuances of trust in the human-AI relationship, we challenge traditional norms and open doors to new possibilities. Listen in as we navigate the digital landscape, highlighting the transformational impact of AI on creativity, innovation, and beyond.

Hosted on Acast. See acast.com/privacy for more information.

In this episode of our podcast, Co-creating with AI, we delve into the concept of AI native apps and their potential for autonomy. We begin by discussing our definition of AI native apps, which revolves around the idea that everything a user can do with a keyboard and mouse should also be doable by AI – effectively opening up new possibilities for app development.

As we explore the notion of AI native apps, we touch upon Adobe Firefly as a prime example of a feature within Adobe Photoshop that brings AI native capability. We contemplate how AI can become autonomous agents, working directly with the backend of an app and executing complex tasks based on given prompts.

Rasmus mentions an example involving a game engine like Roblox or Unity, where an AI-driven workflow could help in expanding the game's development process. We debate on whether autonomy is a requirement for AI native apps, and how this autonomy could enable co-creation with AI.

Towards the end of the episode, we establish a connection between AI native and autonomous AI, concluding that combining these two concepts could lead to purposeful co-creation with AI. The idea of purposeful autonomy resonates with us, and we propose discussing this further in a future episode.

#business #AI #cocreating

Hosted on Acast. See acast.com/privacy for more information.

In this episode of Co-creating with AI, we discuss the current state of AI and focus on how AI can be integrated into different aspects of our lives, especially when it comes to repetitive knowledge work. We explore the idea of AI-native applications, highlighting how the development of AI-centric user interfaces and experiences can revolutionize the way humans work and co-create with AI.

We begin by identifying the capability of AI models, such as language models, which have made it possible to automate repetitive knowledge work. We emphasize that these AI models are more than just assistants, but rather work with humans in organizing and creating knowledge. We also touch upon how some interfaces, such as Adobe Firefly, have started to implement AI in their specialized tools.

By considering the state of AI as an assistant, we ponder how it can be optimized for co-creation between humans and AI. We conclude that to do this, AI must be part of an interface that is both adaptable and suited for organizing and creating content. Our discussion leads us to introduce and explore the concept of AI-native applications, and how rethinking UI and UX for this new world of AI could lead to new collaborative designs and use cases.

Hosted on Acast. See acast.com/privacy for more information.

In this episode of Co-creating with AI, we explore the long-term vision of Multiply and how it aims to fulfill unlimited human potential, both for individuals and humanity as a whole. We discuss how Multiply is built on the foundation of a global collaboration graph, where every person and AI can co-create together, breaking down the silos of traditional collaboration tools.

We express our excitement for the multimodal and multi-model capabilities in Multiply, allowing users to work with multiple AI models and various formats such as text, images, audio, and video. With Multiply, we envision a future where we work seamlessly with AI agents as co-workers rather than just assistants. In this collaborative environment, repetitive and mundane tasks can be automated, freeing up our time to focus on more meaningful work.

Moreover, we highlight the importance of the Multiply App Store, allowing the community to easily build and share apps within the platform. This not only encourages collaboration among users but also provides a channel for AI agents to contribute by autonomously creating apps and sharing them with the community.

By the end of the episode, we find ourselves contemplating a future where AI automates its repetitive work, further proving the potential of Multiply in revolutionizing the way we work and collaborate.

Hosted on Acast. See acast.com/privacy for more information.

In this episode, we explore the intriguing world of data ownership and its implications in the age of AI. We begin by discussing the story of the Narrative Clip lifelogging camera and how much valuable data it generates. As we dive deeper into the topic, we find ourselves questioning who owns the data, especially when it comes to AI models being trained on our personal information.

We ponder the legal aspects of data ownership and how laws like GDPR attempt to regulate it. However, we also realize that these legal frameworks might not be well-adapted to the rapidly evolving landscape of AI technology. We discuss the need for a societal framework that acknowledges the value of data and enables individuals and corporations to benefit from it.

The conversation then shifts to the open-source movement in AI and how it democratizes power in the tech world. We marvel at how open-source AI models are competing with closed-source models, even when it comes to training large-scale models with massive computational resources. We also touch upon the leak of Meta's AI models and how it further spurs the open-source AI community.

We then consider the value of different types of data, from health data to online consumption patterns. We imagine a world in which AI models have access to brainwave data, enabling them to understand and potentially manipulate

Hosted on Acast. See acast.com/privacy for more information.

In this episode, we dive into the fascinating world of real-world AI use cases and explore stories of how AI has played a major role in various situations. We discuss the benefits of co-creating with AI, and how it can be an essential tool to enhance productivity and creativity both at a personal and professional level.

We share stories of companies using AI to improve their decision-making processes and even replace product managers. By feeding survey results and interview transcripts into AI models, companies can create product plans that can drive their business forward. We also delve into our own experiences of using AI in our work, such as refining strategies and architectures through iterative interactions with AI models.

Our conversation takes a turn towards the ethical considerations of using AI in the workplace, as we discuss the potential implications of training AI models on individuals' data. We ponder whether data ownership and personal AI rights should be a part of future employment negotiations, and the potential value of private data in the age of AI.

We wrap up the episode by highlighting the importance of data ownership and AI's role in driving value from personal and company data. We encourage our listeners to share their own stories and experiences of co-creating with AI, and invite them to join us in future episodes as we continue to explore the world of AI.

Hosted on Acast. See acast.com/privacy for more information.