AI: Weekly Summary. RTZ #934

1. OpenAI launches GPT 5.2: OpenAI quickly followed its ‘Code Red’ last week with ChatGPT 5.2 this week. The release is part of ‘Project Garlic’, with future enhancements to GPT and ChatGPT 5.1 designed to counter Google’s Gemini 3. Dubbed ‘it’s best model yet’, the release highlights ‘its smartestAI model yet’. Highlights include performance gains across writing, coding and reasoning benchmarks. Also promised is ‘substantial improvement around hallucinations’ in ChatGPT, up to 38% less than with ChatGPT 5.1. The new release is being rolled out to ChatGPT subscriber and API customers. Expect the rapid improvements race to continue industry wide going into 2026. More here and here.

   

2. Disney vs Google while partnering with OpenAI: Disney plays defense and offense in AI this week with a ‘cease and desist’ letter to Google/YouTube, while partnering with OpenAI around its iconic global content library. Particularly for OpenAI’s super-successful Sora offerings, and presumably its text to image products as well. The deal also calls for a billion dollar equity investment into OpenAI. It was a deal three years in the making. The actions against Google seem to be a negotiation marker for further discussions and negotiations. The moves echo Viacom’s offense and defense moves against Google/YouTube almost two decades ago. Disney’s OpenAI deal calls for detailed processes and guardrails in the use of its content for AI applications and services. Expect these actions to be a continuing set of templates for the industry at large throughout this AI Tech Wave. More here.

   

3. Nvidia gets China green light from US: The US/China AI chip tussles continue with the administration granting Nvidia the permission to sell its H200 AI chips into China. While China’s government tries to figure out if they want to allow such a deal. The moves mark the ongoing balancing act Nvidia founder/CEO Jensen Huang is trying to negotiate to keep its China market open, while working with the Trump administration and US China hawks on their concerns over AI chips to China. The moves also expand to other US AI chips companies like AMD and others. And like the previous set of US permissions, call for a 25% take rate by the US government on Nvidia and others sales into China. That compares to a 15% take rate a few months ago. These developments continue the US/China AI/Tech Trade Kabuki dance into next year. Nvidia and other US tech companies do what they can and must through these geopolitical machinations. More here.

   

4. Debate beyond LLM AI Research: The growing schism amongst AI Researchers around LLM AI directions continues, as evidenced in a leading AI conference this week. The debate centers around the need for meaningful new AI Research innovations to go beyond the current AI Transformer/LLM Scaling approach in vogue today. Prominent amongst these voices are Ilya Sutskever, former OpenAI founder, Yann LeCun, former AI Chief Scientist, and Andrej Karpathy, former head of AI at Tesla. There is growing acknowledgement that better approaches in particular are needed around ‘World Models’, for AI products and services to better deal with ‘Physical AI’ in particular. This of course includes the need for synthetic data and other AI approaches for humanoid robots and fleets of self driving cars. These debates and developments are racing ahead both in the US and China, the other major global AI market. More here.

   

5. Apple’s people problems & AI: Apple saw more concerns over its senior people leaving, following Meta’s recent poaching of its Chief Human Interface Designer. And the departure of its AI head John Giannandrea (JG). Complicating things are questions around CEO Tim Cook’s tenure at Apple now that he’s 65. The recent departure concerns were around its Chip future Johny Srouji, whose team champions Apple’s differentiated ‘Apple Silicon’ strategy. The concerns continue despite Srouji’s reassurance he’s not leaving any time soon. Apple continues to enjoy some major opportunities around AI in my view, despite these ongoing people problems. The company continues to have a wide and deep bench of people and technology ecosystems. More here.

   

Other AI Readings for weekend:

1. AI Bubble Debate continues. More here.

   

2. State level AI regulatory moves continue as White House signs AI anti-state regulations Executive Order. More here.

   

(Additional Note: For more weekend AI listening, have a new podcast series on AI, from a Gen Z to Boomer perspective. It’s called AI Ramblings. Now 31 weekly Episodes and counting. More with the latest AI Ramblings Episode 32 on AI issues of the day. As well as our latest ‘Reads’ and ‘Obsessions’ of the Week. Co-hosted with my Gen Z nephew Neal Makwana):

Up next, the Sunday ‘The Bigger Picture’ tomorrow. Stay tuned.

(NOTE: The discussions here are for information purposes only, and not meant as investment advice at any time. Thanks for joining us here)