CPO delay?

Or something else?

I’ve been watching Fintwit attribute the Semiconductor/AI space selloff recently to a Semi-Analysis article discussing how CPO is being delayed, and or not ready for prime time. While others rightly point out that even if that were the case (I have no idea), NPO still benifits’ and is here now.

But what if it were something else spooking the market? I have been watching various headlines over the past months, and the amount of planned datacenters in the US that are either postponed, or cancelled out right is starting to be a meaningful number.

Why would a hyperscaler request a halt when they publicly keep talking about an insatiable demand? What if the math behind Token use (until it gets commoditized) has enterprises finally pushing back?

Why would banks balk at a $6B loan secured by $60B in OpenAi stock?

The push back of ‘Not in my backyard’ grows louder

Now you won’t hear about this from any of the Semi and memory players, this is because in Semiconductor land, double and triple ordering to get your share of a product has a long history ….

Then there’s the past times Semi’s grew over 40% yoy….what happened the following year?

What’s with the performance of all the BESS stocks which were supposed to benefit from a rebuild of the grid, and an insatiable appetite from data centers? What are they telling us?

Though I have not sold my AI/Semi/Optical stocks, and think we are in early innings know just how eerily similar the AI buildout is to the dot.com bubble pop….The reason why there were a gazillion CLEC’s, and the Ascend communications, JDS Uniphase, Sycamore Networks, and Cisco’s of the world were trading at such high multiples was the general consensus at the time. Everyone knew that everything would need to be ported from copper to fiber (Sound familiar?), and the thinking (at that time) was that it would happen overnight (Verizon and AT&T as examples would need to replace every Cat 5 switch with a Softswitch…..every inch of copper on the telephone poles would need to be replaced with fiber. This did all ultimately happen, but it did so over decades as normal capex cycles dictate. This is what imploded the NASDAQ - it wasn’t that the buildout never happened - it’s that it didn’t happen in the timeframe that EVERY tech company and sear predicted.