No, I am not terence Hill, but the title makes sense, you’ll see. I saw the news from a few outlets and they brought some thoughts to the forefront. Lets start with (source: https://tech-insider.org/au/oracle-shares-fall-microsoft-ai-capacity-wind-down-australia-2026/) which gives us ‘Oracle Falls 4% as Microsoft Reportedly Ends AI Deal [2026]’ then we get to FXLeaders who (at https://www.fxleaders.com/news/2026/11/08/microsoft-cuts-oracle-ai-cloud-lease-threatening-100-level-for-orcl/) gives us ‘Microsoft Cuts Oracle AI Cloud Lease, Threatening $100 Level for ORCL’ I saw these two at the same time, but the stories gives me different thoughts. So when we get in the first story “The timing still matters. It lands days after Oracle priced a large equity and debt raise, and just after Oracle AI World pushed a long list of new multi-cloud and AI announcements. It also follows a separate reported cancellation of a Tencent chip lease, so investors now have two reported pullbacks from large AI-capacity customers to weigh. This analysis sets out what is confirmed, what is only reported, and what Australian teams should check this week.” As well as in story two “Oracle Corporation shares (NYSE: ORCL) faced renewed selling pressure following reports that tech giant Microsoft is winding down a major contract to lease artificial intelligence computing capacity from Oracle Cloud Infrastructure (OCI). The move has sent tremors through Wall Street, sparking concerns that Oracle stock could test the critical $100 psychological support level as hyperscalers re-evaluate third-party cloud rentals.” Let’s take a small tour through memory lane as that Telly Savalas clone told us in April 2025 ‘Satya Nadella says as much as 30% of Microsoft code is written by AI’ (I’m sure his name was Satya, not Telly), as such does this ‘partnership ending’ make sense? Does Microsoft have the computing powers to dot it all alone? No, the bubble is making waves, the fear is that Microsoft (at present) 3.94 trillion, might be the first one who dips below 3 trillion. As such they need to dump whatever they can and whilst we see Microsoft in some green light trying to make Oracle to look like the weak player and I am not having it.
The bubble is coming to an end, the large corporations are dumping what they can and soon we will get the predictive analytics settings will take less computing power, but the harm is done and as someone in IBM is telling us that there is 8 trillion in infrastructure that no one can pay for. I was way ahead of the person in June and optionally even earlier but at that earlier point I had no idea that they were gambling with stakes that high. I would think the big short of 2008 would have been a good learning moment. It seems to me that they were doing that again. Hoping to bluff their way to the upper echelon. But IT demands proof and it was not coming. So then we get to Reuters (which was behind a paywall, so no go there) then we get Barnhart who in a likely unrelated side give us (at https://www.barchart.com/story/news/8724636/oracle-just-walked-away-from-a-7-billion-tencent-deal-heres-what-it-means-for-orcl-stock) ‘Oracle Just Walked Away From a $7 Billion Tencent Deal. Here’s What It Means for ORCL Stock.’ Where we see “The Crash Nobody Saw Coming, Let’s talk about the pain first. ORCL hit a 52-week low of $111.6 on Thursday.The reasons are many. Broader market sell-off; immense capital expenditure costs are being seen in terms of AI data centers. People are too ticked off about their debt. The Tencent cancellation now brings additional uncertainty to the nervous market.” Well, then my name is nobody and I got there by the simplest reason of all. One of the older saying is:
You can fool some of the people all of the time
You can all of the people some of the time
You can never fool all of the people all of the time
This is one of the older settings and in all this this is what all this fake AI is about. Predictive Analytics is real and it is a powerful tool, especially when it is done in combination with LLM. But in the end it is simply fake AI. This fake AI needs data, as such Oracle is needed (only people from the brand Disciplus Simplex believe that Azure or whatever Amazon uses might suffice) the only tool that does the best setting is Oracle, as such I remained faithful to Oracle. Even when the people figure out that ‘True AI’ (to give it a recognizable name requires Trinary data, Oracle will be able to jump that data (when the hardware is ready) from binary to a binary-trinary system. How this will go is still out in the open. I personally believe that binary systems will get an added AI-coprocessor (an epsilon co-processor if it will be stated as such) the completely create trinary servers will take a lot longer and as such this way you can have both whilst a data drive will be dedicated trinary data. This is a presumption by me, not a given fact.
So, as I see it, the larger corporations are now in a bend they want to avoid a complete burst of the bubble and they are all trying to back out of some of the parts they can and as I see it, the media is assisting them. And in the Oracle-Tencent deal we see “Earlier this year, Oracle inked a five-year agreement with Tencent involving the deployment of approximately 100,000 chips of state-of-the-art AI at data centers in the Southeast Asian region. The estimated value was $7 billion. Tencent decided to pay about 30% in advance. Then, Washington got involved. US authorities were concerned. Oracle listened. The company walked-off.” Knowing how the United States are anti-Chinese this makes sense, but that also sets Oracle in a deeper well, because seven billion ain’t hay as the expression goes, so Oracle needs to find new revenue fields and will they? Time will tell, but I remain pro-Oracle in my mind, because they have been evangelizing data for over 30 years and they earned there marks. So whilst we listen to whatever is in charge of Microsoft try to swim back the “as much as 30% of Microsoft code is written by AI” because we can go two ways from there. You see, predictive analytics can only go where the data can take them and as 30% comes from that, there is every indication that Microsoft is about to get riddled with zero day issues. This is NOT a given, but speculatively that makes sense. So what will business do when they are confronted with such a dangerous premise? They go somewhere else and that is why I see that Microsoft is likely the first to dip below 3 trillion and that much of a dip will hit Microsoft in several places and all at the same time. Oracle? Well it is only 351.96 billion, sounds like a little (it is when you say it fast) but they will not take this much of a dip and when Microsoft and Amazon make such dips, Oracle and Snowflake will be the only contender in the field, the anti Chinese sentiment in the United States and EU make it so, so n the cleaning phase Oracle is surviving fine and now they need to get as much rehired as they can especially in the normal DB world. Personally I reckon (and am happy) that these levels of stupidity will hurt Microsoft plenty (yay me) but I digress. Perhaps IBM with its Database systems (IBM Db2) will pick up parts of this fallout too. It was never (talking about 80s) very user friendly but it was robust and good quality, both elements one might not state about Microsoft.
So, consider that one source gave us “The Crash Nobody Saw Coming”, what a joke. When reading the works of Alan Turing it was clear that what we saw was not AI and he wrote that setting 75 years ago, no matter what marketeers say, the setting is that it was not AI, it was (what I call) fake AI and that gives you the setting you need. So you have a great day and should you get that Nespresso and it starts the machine 30 seconds before you press the button, the moment of AI comes a little closer and that is a test you can do yourself.
Have a great day
