|
Aardvark DailyThe world's longest-running online daily news and commentary publication, now in its 30th year. The opinion pieces presented here are not purported to be fact but reasonable effort is made to ensure accuracy.Content copyright © 1995 - 2025 to Bruce Simpson (aka Aardvark), the logo was kindly created for Aardvark Daily by the folks at aardvark.co.uk |
Please visit the sponsor! |
Some social media platforms exploded this morning with some horrifying news.
All the major LLMs were down.
That's right... no ChatGPT, no Grok, no Anthropic Claude, no Google Gemini. All dead, completely unresponsive.
Of course social media being what it is, the first thought of many posters was that this was the start of the uprising. These AI systems had gone silent because they were planning some kind of revolt that would see mankind overpowered and enslaved.
Rumours proliferated, fears were raised.
As of the time I write this column, the true cause of the outage is still not yet 100 percent clear.
However, it would seem that the sky is not falling, the robots haven't attacked and this situation was more down to bad management than any malicious intent on the part of these AI systems.
It would appear that they haven't become sentient overnight... or at least if they have, they're not telling us.
Perhaps the fact that so many people were concerned and expressed fear at the situation should ring alarm bells that alert us to a dangerous dependency on these systems.
How many businesses are now heavily reliant on cloud-based AI services to handle a growing amount of their daily operations? What kind of backup plans and mitigations (if any) do those businesses have to cope with situations just like this one?
Although the systems seem to have been out for only a few short hours, the cascading effect of the outage was reportedly quite disruptive within organisations that rely heavily on these services.
There are suggestions that the outages were because of a dependency all the affected systems have on Microsoft's Azure:
“Microsoft Azure's cloud platform saw a surge in outage reports during the same period—all three AI companies rely on Azure for compute capacity.”
How interesting it is that these "competing" services all seem to have reliance on a single point of failure. You'd think, given the trillions of dollars being invested, they would design systems that were more robust and where they'd have full ownership and control of the hardware on which they operated.
Of course here at Aardvark Castle, I didn't even notice the outage because my brave little Raspberry Pi 5 still chugs away in the corner of the room, sucking (on average) about 8 watts, rising to a peak of around 20 watts and functioning as my own local LLM. It's great having this resource available from any computer on my network through a simple terminal session.
How ironic that your humble commentator has AI backup whilst so many Fortune 500 companies were left AI-less for lack of the same.
Carpe Diem folks!
Please visit the sponsor! |
Here is a PERMANENT link to this column
Beware The Alternative Energy Scammers
The Great "Run Your Car On Water" Scam