AI: Weekly Summary. RTZ #928
...week ending December 5, 2025
OpenAI’s ‘Code Red’ on 3rd ChatGPT Anniversary: OpenAI ironically had an internal ‘Code Red’ moment three years after Google’s similar ‘Code Red’ at the end of 2022. It of course occurred three years after the launch of OpenAI’s ChatGPT. The general drivers this time seems to be Google’s success with its recent Gemini 3 launch. As well as continued progress by US and Chinese LLM AI companies vs OpenAI’s core products. OpenAI’s Sam Altman is apparently postponing other initiatives at the company like AI Ads, AI personalization and AI ecommerce initiatives, amongst others. OpenAI ‘Garlic’ iteration for GPT in the coming months is expected to be a response to Gemini 3, along with other iterations expected as early as this month. More here and here.
Anthropic’s Alternative AI Path: Anthropic is having a smoother path on the enterprise track than OpenAI with its consumer and other AI ambitions. This is a distinction Anthropic founder/CEO Dario Amodei is increasingly making at public forums. And its potentially earlier path to profitabiity. Especially given OpenAI’s Code Red and other challenges. It’s a tactically smart strategy given that both companies are targeting IPOS next year and beyond. Especially as their valuations cross $300-350 billion for Anthropic and over $500 billion for OpenAI. That Anthropic now counts Microsoft and Nvidia as investors along with Amazon and Google adds to the contrast. Especially given OpenAI’s slow separation from its iconic Microsoft partnership. More here.
Amazon AWS ‘Re-Invents’ AI Services: The largest US AI Cloud Data Center company had its annual Re:Invent Developer conference, where it highlights a range of broad and deep AI services for businesses around the world. WIth its recent partnerships with LLM AI companies beyond Anthropic now including OpenAI, Grok and others, as well as its own AI models, Amazon AWS is now on track to bring a robust array of AI products and services across a range of AI chip infrastructures beyond Nvidia. Amazon also now has the ability to offer these services across multiple hardware and software environments, with a varying range of customization capabilities as needed by its enterprise customers. Amazon AWS remains well positioned vs its peers Microsoft Azure and Google Cloud in particular. More here.
Meta Poaches Apple Design: Meta poached Apple’s Chief Designer to set up its own AI Wearables focused Design unit within its AI Superintelligence Reality Labs. Founder/CEO Mark Zuckerberg announced the moves on Threads. This establishes Meta as a player at scale in AI Devices, following OpenAI’s $6.5 billion acquisition of ex-Apple Designer Jony Ive’s company a few months ago, to design its own portfolio of AI Devices starting next year. This now means that Apple, along with Google and Amazon with their earlier base of ‘smart devices’ being revamped for LLM AIs, will see robust competition from Meta and OpenAI, as well as startups from the US and beyond. Meta continues the most well-financed and laser-focused initiative relative to its peers. More here.
US Startups going China open source: More US startups and companies are now using open source LLM AI models from China in addition to various offerings by US companies. The drivers are price, more customizability, and sharply higher capabilities in AI Reasoning relative to their US counterparts. Alibaba with its Qwen models in particular have made notable headway, along with over half a dozen other companies large and small. This is occurring as the former lead open source US company Meta, is increasingly focused on hybrid approaches. Other US open source startups are emerging with Reflection AI being a recent example. But the Chinese offerings are becoming competitive at scale for a wide range of AI applications. Not just for US tech startups but for a host of US businesses large and small as well. This of course has implications on the regulatory front given the ongoing US/China ‘AI Space Race’ in progress, driven by geopolitical imperatives. More here.
Other AI Readings for weekend:
New York Times sues ‘AI Bad Boy’ Perplexity . More here.
Softbank’s Masayoshi Son’s Grand ‘Project Crystal Lands’ US AI Plans. 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 31 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)
That bit about US startups using Alibaba's Qwen models is actually a big deal. When you can get comparable performance at better prices with more customization, the vendor choice becomes obvious regardless of where the tech comes from. Gonna be interesting to see how regulators try to square this with the whole AI race narrative.