first-hand human knowledge
We read the podcasts, essays and interviews — and hand you the arguments: who claims what, who rebuts, and the original voice one click away.
The hosts argue that despite numerous crises including pandemic, inflation, wars, tariffs, and banking failures, the AI trade has single-handedly kept the market at all-time highs.
Nathan Sobo argues that with the rise of AI agents, conversations now initiate the coding process, making code a product of those conversations rather than the starting point.
The host introduces the black market for AI tokens, where Chinese sellers offer API access to services like Claude and OpenAI for a fraction of the official price, with some AI companies reportedly losing millions per month.
Brett Adcock lays out his thesis that AI will evolve along two parallel tracks: humanoid robots for physical work and a 'Jarvis-like' personal AI that knows your memories, accesses your accounts, and handles tasks autonomously.
While personal boundaries are important, workplace systems and employment conditions ultimately determine whether developers can actually stop working on time.
Survey data shows 59% of developers feel their skills are diminishing and 54% enjoy coding less, with the same people reporting both issues.
Brett Adcock predicts that AI will dwarf the internet in scale and impact, citing the speed of progress in deep learning as evidence that this trend is accelerating faster than anything he saw in 15 years of software and internet work.
Yana Welinder demonstrates that using AI tools like Codex and ChatGPT allows her to go from hand-drawn sketches to a live e-commerce site with 3D-printed garments, all without technical staff.
Massa argues that simply giving teams ChatGPT or Claude without restructuring leads to no real efficiencies; companies must rebuild APIs and systems specifically for agent use.
AI tools shift work from coding to reviewing and decision-making while creating pressure from managers who expect increased output without understanding the actual workflow changes.
Doug Turnbull's approach: have the model emit imagined tags unconstrained by the existing vocabulary, then use vector embeddings against the corpus to find the concrete tags closest to what it invented.
Adcock describes AI's two future paths—humanoid robots for physical tasks and a digital AI partner that knows everything about your life, has perfect memory, and acts as a superhuman assistant always with you.
The paper breaks the assumption that lab reasoning traces are secure. It shows encrypted reasoning blocks can be decoded and replayed, ported to weaker models to transcribe, improving open models — and leaking personal data from publicly shared traces.
Kavak built sales agents rather than support agents, and these agents now convert over 2x better than human teams while tripling NPS scores.
Silber argues that we are all extremely early in the AI design process, making now the best time in history to enter the field.
Dario Amodei attributes public negativity toward AI to a pre-existing crisis of trust in institutions, rather than to risk warnings from AI leaders.
Ian Silber argues that the current technological moment, particularly with AI, presents an unprecedented and optimal environment for product designers to create and innovate.
Adcock argues that pursuing hard problems is strategically superior because they attract less competition, better talent, and investors seeking binary payoffs. He illustrates this with humanoid robots vs. robot dogs — humanoids may be only 3-4x harder but could yield a million times higher ROI.
Brett Adcock frames his companies as being at a critical inflection point where the outcome is binary — either robots scale or they don't — and all his energy is focused on making them 100x to 1000x bigger.
Nathan Sobo introduces Delta DB, a version control system that records work as it unfolds, keeping every change connected to the conversation, enabling conflict-free work trees and a new way to collaborate with agents and teams in real-time.