AI: A look back at 2025 in AI land. RTZ #942
The Bigger Picture, December 21, 2025
It’s that time of year in 2025, a look back on the year for this AI Tech Wave. It builds on the one I did for 2024 last year, and the expectations for 2025 published this time last year. A brief review of 2024 is the Bigger Picture I’d like to focus on this Sunday.
It’s been a busy year for all things AI with tons of surprises up and down the AI Tech Stack above. I’ve been discussing them daily here on AI: Reset to Zero (RTZ), with total posts up to #942 this Sunday. And 33 weekly AI Ramblings podcasts thus far with my Gen Z software engineer nephew, Neal Makwana.
The biggest development of course has been the global mainstream acceptance of AI Infrastructure investments in Boxes 1 through 5 above, with investments per big tech company approaching $100 billion per year, with multi-year budgets of hundreds of billions per year per most of the Mag 7s.
The WSJ has a good piece on how the AI industry is looking back on the year at the NeurIPS AI industry Conference in San Diego this week.
If one looks just at OpenAI, its ambitions on AI Data Center and Power investments approach a trillion and a half dollars, with aggressive plans to ramp up revenues over the next few years to justify that level of financing.
Taking a step back on the year, there are five big ‘Surprises’ for the year from my perspective. Here they are in no particular order, along with my AI:RTZ post that discussed each one in detail.
-
OpenAI’s Extraordinary Gauntlet: AI-RTZ # 895
This of course has been a key AI narrative of the year, with Sam Altman and team driving OpenAI to new heights almost ever week, notwithstanding the internal ‘Code Reds’.
-
Google’s AI Comeback: AI-RTZ #893
Google CEO Sundar Pichai showed the rest of the world he was a ‘war time’ CEO after all. Just a month after OpenAI’s ChatGPR launch on November 30, 2022, he declared his own internal ‘Code Red’, and turned Google around in a front-line AI leader in three years. Now Google is ready for the AI ultra-marathon ahead.
-
Meta and Zuck’s AI Superintelligence move away from Llama: AI-RTZ #891
A big surprise this year of coruse was Meta founder/CEO Mark Zuckerberg’s decisive move away from its then world-leading open source Llama LLM AI, and an unprecedented industry wide AI Talent drive to revamp its AI strategy going into 2026. Hundreds of billions in AI investments notwithstanding.
-
China taking the Global Open Source Title: AI-RTZ #823
China surprised the world post its DeepSeek surprise early this year, with its AI companies all switching to open source models in rapid order. And then leading the world on that front by year end. Not to mention tech incumbent Alibaba taking the lead on all that in the same time.
-
World Wobbles through Tariffs, so far: AI-RTZ #684
The abrupt US ‘reciprocal Tariff’ initiative in April upended most global trade and supply chains across hundreds of countries. Not to mention the core US/China trade kerfuffles hinged on mutual geopolitical issues. For now, 2025 is ending with both countries balancing and ‘threading the needle’ on these issues, for the most part. And Nvidia’s Jensen Huang executing well on his own balancing act to boot.
There were of course many other big things in AI this year. But the five above stand out for me the most. I also discuss them in more detail on my weekly AI Ramblings podcast with my Gen Z nephew Neal Makwana.
It’s Episode 34 below, a review of AI in 2025 and its surprises.
We also go into more detail on the most ‘overhyped’ AI items of the year. And our favorite AI products and services in 2025. Would add all that to the Bigger Picture this Sunday. Worth a watch through this holiday period.
Best wishes for Christmas and Holidays ahead.
Next Sunday, will discuss the ‘look ahead’ to 2026 this AI Tech Wave. 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)
