Pages

Saturday, February 21, 2026

Focus: Black Swans -- February 21, 2026

Locator: 50021BLACKSWANS.

Focus, link here: black swans. 

Updates

Later, 2:06 p.m. Central time: on January 30, 2026, I asked whether Satya Nadella's job was on the line. It would take a lot for the board to remove Nadella and I don't think his job is on the line, but once your business model -- a subscription model -- comes into question, and your most visible product -- Copilot-- appears not to be a hit among consumers, a CEO has a lot of headwinds to weather.  

Original Post 

AI investing

  • AI: too much debt in the short term spooks investors;
  • AI: upsets need for security software.

AI prompt: AI investing. Somehow analysts suggest AI eliminates need for security software companies. Is that why software companies plummeted in the market this last week? 

Reply

The bigger question has become: is AI going to be a much bigger "disruptor" than previously "believed"? 

Another question: does a "shoot first, ask questions later," offer some interesting investing opportunities? 

By the way, here's a huge counter-argument: 

  • folks in general do not trust AI / chatbots"
  • "everyone" agree AI / chatbots are somewhere between 90% ad 95% accurage
  • do you trust AI to find security risks? 
  • to what extent can companies like CrowdStrike, Okta, and Cloudfare use this to their advantage?

But I think there was more to it that that

AI prompt: Per your earlier reply, the significant decline in software and cybersecurity stocks around early-to-mid February 2026 was largely driven by investor panic over the disruptive potential of Artificial Intelligence. Separate from that, I believe, was the cocern investors have regarding Microsoft's 1) subscription model; and, 2) its poorly accepted Copilot. Your thoughts? 

Reply (and it is a long reply): 

AI prompt: And finally, one last question along this line: it appears to me that agentic AI may be more disruptive than previously thought. Not too long ago, the question was whether AI / agentic AI is a "bubble"? Now it appears that not only is AI / agentic AI NOT a bubble but it may be much more disruptive than previously "believed." Again, your thoughts. Looking for an AI inference reply.

Response: after three attempts to get a narrative reply from Google, I gave up. Google only provided links to articles, the old Google search method. These were the first two hits: