MoE, Mavericks & Missteps: What You’re Missing About the AI Race
Failed to add items
Add to basket failed.
Add to Wish List failed.
Remove from Wish List failed.
Follow podcast failed
Unfollow podcast failed
MoE, Mavericks & Missteps: What You’re Missing About the AI Race
-
Narrated by:
-
By:
About this listen
In this episode, we unpack the evolving landscape of generative AI — from the delay of GPT-5 and the release of OpenAI’s O3 and O4 Mini models to the explosive rise of Chinese innovations like DeepSeek and Mistral. We dig into why calling GPT-5 the 'holy grail' of AI is a flawed narrative, and how the global AI race is being reshaped by open-weight models, MoE (Mixture of Experts) architectures, and aggressive Chinese strategies.
We also explore the growing buzz around 'white coding' — where non-programmers use AI to generate code — and the infrastructure shocks that follow. Why are API bills skyrocketing? What happens when blindly deployed AI code meets real-world production? From misleading claims to misunderstood models, we break it all down.
Tune in as we question hype, bust myths, and offer a grounded take on where AI is really headed.