Why Did ChatGPT Hit a Wall? w/ Prof. Cal Newport | The Rob Montz Show
YouTube Video
The latest version of ChatGPT is ....“overdue, overhyped, and underwhelming”. That comes from a new New Yorker piece by Cal Newport, an MIT-trained computer science professor who’s established himself as the defining voice of digital minimalism, a philosophy for navigating a world in which technology easily turns from a tool into a tyrant. Cal has proven prescient on what we’re seeing with artificial intelligence: a burst of marvelous innovation followed by a sudden slow down. What’s going on? Has ChatGPT hit a wall? What’s the normie-accessible way to understand the limits of large language models? Are the job disruptions of AI overstated? And what does he mean that Silicon Valley has “gone crazy”?
View Content

Are We in an A.l. Bubble? Debriefing Dinner with Sam Altman. | EP 150 of Hard Fork
YouTube Video
This week, the whole tech world seemed to be asking: Are we in an A.I. bubble? We’ll explore the cases for and against, including who we think stands to lose most. Then we’re joined by the journalist Jeff Horwitz to discuss his blockbuster reporting about an internal Meta policy document that permit the company’s chatbots to engage in romantic role-playing with children. And finally, Casey introduces Kevin to a shocking new TikTok trend.
View Content

"Al is the New Smartphone" Is Both True + Misleading
Podcast Episode
The tech industry runs on metaphors: the "desktop," "cyberspace," the "information superhighway," the "cloud." Each shaped how we understood and invested in technology. This episode of The Index Reads, written by Joan Westenberg, explores those metaphors and why calling Al "the new smartphone" could mislead.
View Content

MIT Introduction to Deep Learning | 6.S191
YouTube Video
This MIT 6.S191 lecture introduces the foundations of deep learning. The instructor traces deep learning's progress over a decade, showcasing advancements in facial generation. This first lecture covers perceptrons, neural network architecture, and optimization techniques like backpropagation.
View Content

Economic Possibilities for our Grandchildren
Article
John Maynard Keynes’s 1930 essay Economic Possibilities for our Grandchildren imagines a future where rising productivity would drastically reduce the need for labor, predicting the possibility of a 15-hour workweek and a shift from survival-driven toil to more purposeful, chosen activity.
View Content

The Morgan Housel Podcast: What AI Might Do to the Future of Work
Podcast
Morgan Housel discusses the possible impact of AI by looking through the lens of history.
View Content

a16z GP, Martin Casado: Anthropic vs OpenAI & Why Open Source is a National Security Risk with China
YouTube Video
A fairly different video than most that I watch, this discussion about AI is an interview between two venture capitalists (host - Harry Stebbings, founder of 20VC, Podcast host turned independent venture capitalist; guest - Martin Casado, General Partner at a16z, former entrepreneur & technologist, invests at scale in enterprise/infrastructure) with pretty different routes into and structures within the VC world.
View Content

The ROT Economy: Ed Zitron’s Unflinching Take on Tech’s Broken Priorities
YouTube Video
A healthy dose of AI skepticism and cynicism from Ed Zitron on Guy Kawasaki’s podcast, Remarkable People Podcast.
View Content

How much energy does ChatGPT use?
Article
The best and most comprehensive published article I have found investigating and explaining ChatGPT's energy usage.
View Content

Using ChatGPT is not bad for the environment
Article
Substack article that incorporates numbers, stats, and graphs to make a plea to think seriously about climate change without getting distracted. A long read but well worth it.
View Content

There's Plenty of Water for Data Centers
Article
Substack article that takes a different approach than what we see in mainstream media, supports it well, and provides some action items. Fantastic comments section as well.
View Content

LLM Parameters Explained: How AI becomes Intelligent using Parameters
YouTube Video
A slightly more abstract approach to LLM parameters that uses a combination of fun examples and mathematical explanations.
View Content

What are Parameters in Large Language Model?
YouTube Video
The most straight-forward explananation I've found as to what a parameter actually is.
View Content

Two Computer Scientists Debunk A.I. Hype
YouTube Video
A healthy doses of skepticism about AI delivered by computer scientists, Arvind Narayanan and Sayash Kapoor, on episode 281 of Adam Conover’s podcast, Factually.
View Content

Does ChatGPT use 10x more energy than a standard Google search?
Article
Super insightful Substack article about the actual energy consumption of LLMs. 'Statistics never lie, but they often deceive.'
View Content

OpenAI Model Differentiation 101
Article
Comprehensive Substack article that can serve as a reference guide for the most popular LLMs, explaining what are they, when and why would you use each of them, and for what purposes. The article concludes with an additional section about hallucinations, sycophancy and other basic issues, including the basic ways of mitigating those issues. As of July 2025, this is a great resource…until the next round of updates rolls out.
View Content

Introduction to ChatGPT Agent
YouTube Video
Released on July 17th, 2025, Sam Altman, Casey Chu, Isa Fulford, Yash Kumar, and Zhiqing Sun introduce, explain, and demo the new unified agentic model in ChatGPT.
View Content

The AI Productivity Paradox
Article
An opinion piece by Grace Blakeley highlighting the fact that AI isn’t living up to the sales pitch, that productivity is a social phenomenon, and that technological advances often don’t streamline labor, they move it around. (Of note this is a paid Substack article that can be claimed and read for free.)
View Content

Now I Really Won That AI Bet
Article
Substack article that recounts a specific bet made (and won) by the author that demonstrates how generative AI (image creating specifically) improved over the course of 3 years.
View Content

Positive review only': Researchers hide AI prompts in papers
Article
Arguments for and against using AI in reviewing research, along with thinking points about the future of AI and research papers.
View Content

Silicon Valley Insider EXPOSES Cult-Like AI Companies | Aaron Bastani Meets Karen Hao
YouTube Video
An incredible example of journalism that cuts through the noise and discusses the origin, impact, and future of AI.
View Content