Uploaded December 2025 | Updated September 2026, 1 week ago
Launch your site for free at framer.link/WesRoth
Use code WESROTH for a free month on Framer Pro.
Build a site that looks hand-coded. Without hiring a developer.
______________________________________________
VIDEO SUMMARY
In this video, we test the newly released GPT 5.2 Pro and its "extended thinking" capabilities. We push the model to create complex 3D simulations—including a spherical Conway's Game of Life and a destructible city game—in a single prompt. The results show a model that acts less like a chatbot and more like a remote engineer, taking up to an hour to reason through code architecture before delivering a final project.
We also break down the new "GDPval" benchmark. unlike traditional tests, this evaluates AI against human experts with an average of 14 years of experience in fields ranging from finance to mechanical engineering.
The latest AI News. Learn about LLMs, Gen AI and get ready for the rollout of AGI. Wes Roth covers the latest happenings in the world of OpenAI, Google, Anthropic, NVIDIA and Open Source AI.
Emad Mostaque Interview
"No One is Prepared" the next 1,000 days are CRUCIAL
youtube.com/watch?v=07fuMWzFSUw
OpenAI Introducing GPT-5.2
openai.com/index/introducing-gpt-5-2
______________________________________________
My Links 🔗
➡️ Twitter: https://x.com/WesRothMoney
➡️ AI Newsletter: natural20.beehiiv.com/subscribe
Want to work with me?
Brand, sponsorship & business inquiries: wesroth@smoothmedia.co
Check out my AI Podcast where me and Dylan interview AI experts:
youtube.com/playlist?list=PLb1th0f6y4XSKLYenSVDUXFjSHsZTTfhk
______________________________________________
[00:00:00] Intro: 3D Spherical Conway's Game of Life
[00:01:03] GPT 5.2 Release
[00:02:55] Ethan Mollick & Noam Brown on GDP-Eval
[00:03:52] Framer (Sponsor)
[00:05:55] What is GDPval?
[00:14:30] Economic Implications: When AI Outperforms Experts
[00:16:15] Other Benchmarks: SWE-Bench, MATH, & ARC-AGI
[00:18:10] Qualitative Leap: Cap Tables & Project Management
[00:19:00] The Intelligence Curve: Performance vs. Compute Cost
[00:20:40] 390x Cost Reduction in One Year
[00:22:40] Addressing Skeptics: "Stochastic Parrots" vs. Real Utility
[00:26:08] Model Testing
#ai #openai #llm
Launch your site for free at framer.link/WesRoth
Use code WESROTH for a free month on Framer Pro.
Build a site that looks hand-coded. Without hiring a developer.
______________________________________________
VIDEO SUMMARY
In this video, we test the newly released GPT 5.2 Pro and its "extended thinking" capabilities. We push the model to create complex 3D simulations—including a spherical Conway's Game of Life and a destructible city game—in a single prompt. The results show a model that acts less like a chatbot and more like a remote engineer, taking up to an hour to reason through code architecture before delivering a final project.
We also break down the new "GDPval" benchmark. unlike traditional tests, this evaluates AI against human experts with an average of 14 years of experience in fields ranging from finance to mechanical engineering.
The latest AI News. Learn about LLMs, Gen AI and get ready for the rollout of AGI. Wes Roth covers the latest happenings in the world of OpenAI, Google, Anthropic, NVIDIA and Open Source AI.
Emad Mostaque Interview
"No One is Prepared" the next 1,000 days are CRUCIAL
youtube.com/watch?v=07fuMWzFSUw
OpenAI Introducing GPT-5.2
openai.com/index/introducing-gpt-5-2
______________________________________________
My Links 🔗
➡️ Twitter: https://x.com/WesRothMoney
➡️ AI Newsletter: natural20.beehiiv.com/subscribe
Want to work with me?
Brand, sponsorship & business inquiries: wesroth@smoothmedia.co
Check out my AI Podcast where me and Dylan interview AI experts:
youtube.com/playlist?list=PLb1th0f6y4XSKLYenSVDUXFjSHsZTTfhk
______________________________________________
[00:00:00] Intro: 3D Spherical Conway's Game of Life
[00:01:03] GPT 5.2 Release
[00:02:55] Ethan Mollick & Noam Brown on GDP-Eval
[00:03:52] Framer (Sponsor)
[00:05:55] What is GDPval?
[00:14:30] Economic Implications: When AI Outperforms Experts
[00:16:15] Other Benchmarks: SWE-Bench, MATH, & ARC-AGI
[00:18:10] Qualitative Leap: Cap Tables & Project Management
[00:19:00] The Intelligence Curve: Performance vs. Compute Cost
[00:20:40] 390x Cost Reduction in One Year
[00:22:40] Addressing Skeptics: "Stochastic Parrots" vs. Real Utility
[00:26:08] Model Testing
#ai #openai #llm
