Why is everything everywhere suddenly breaking?
We’ve had a recent wave of major technology outages. The Microsoft Outlook really frustrated me (happened on the day I decided to finally tackle my inbox… five hours later I was in a cold sweat wondering if those emails were lost to the cosmos)… Apple services experienced disruptions… Amazon Web Services had problems… banking systems stumbled… airlines couldn’t get planes off (or back on) the ground… and even the National Security Agency reportedly experienced issues with its internal systems (gulp!). Different organizations… different technologies… different failures.
All of this happening while we’re celebrating some of the most extraordinary technological advancements in history.
We’re all watching artificial intelligence accomplish things that seemed impossible just a few years (months?) ago… while struggling to keep some of the most basic infrastructure of our digital lives functioning. Is something bigger happening here (cue The X-Files music)? Are these simply independent incidents that happen to be getting more attention… or are we starting to see the unintended consequences of building an increasingly sophisticated world on top of technology that wasn’t designed to support it?
The technology continues to impress while the experience of transformation is getting lumpier.
When the Federal Aviation Administration grounded flights across the United States for nearly five hours last week, the disruption spread well beyond the American border and some travellers were still trying to get where they were going days later. What caused the problem? Early indications point to new networking equipment interacting with aging telecommunications infrastructure. Investigators are still determining the precise cause, but the irony is thick. Upgrading didn’t just downgrade the system… it took it down for the count. Let’s not let our imaginations wander… there’s no evidence that this was related to AI… a coordinated cyberattack… or that it was connected to other #techfails. We spend so much time wondering whether humans can absorb the speed of technological change… what if the technology we already depend on is struggling to absorb it, too?
What a time to be alive.
I spend a lot of time speaking with business leaders about AI and the next era of technology. And one of the things that keeps coming up is how uneven this transformation has become. On one side, we’re watching AI-first businesses and technologically sophisticated organizations make extraordinary leaps. On the other side, we’re seeing established organizations struggle to move even relatively modest AI experiments into production (even as individual employees experiment with these same tools on their own). The technology is capable… the organization isn’t.
Here’s the story I hear most often (does it sound familiar)…
Boards and executive teams are pushing for AI adoption, scale, productivity and measurable returns (with serious money on the line), while IT teams are managing roadmaps that exceed the resources (and, sometimes, the frontier AI expertise) available to deliver them. They’re responsible for decades of legacy infrastructure, fragmented databases, cybersecurity, compliance and the everyday expectation that everything needs to keep working. Now they’re being asked to introduce an entirely new layer of intelligence into systems that were never designed for anything like it. Add in the human dimension. People who have spent their careers developing specialized expertise are understandably trying to figure out what happens when new technology challenges the value of that knowledge. And being sophisticated in enterprise technology doesn’t automatically mean understanding what’s possible with today’s frontier AI models.
The leaders pushing hardest for technological transformation may not fully understand the systems they’re trying to transform… while the people responsible for those systems may not fully understand the possibilities of the new technology.
It’s all rocks and hard places. A failed experiment in a sandbox might be a useful lesson. A failed deployment in a live environment can have consequences that ripple far beyond the organization itself. Even careful testing can’t necessarily recreate every interaction between new technology and older systems once everything goes live. The FAA incident is a reminder of that reality. While the aviation system experienced an enormous operational disruption, its safety procedures worked. 3500-ish aircrafts already in flight landed safely. It’s a good reminder that there’s a big difference between system failure and a system being incapable of managing failure. I don’t see any reason to become pessimistic about innovation.
But I do wonder if we’re spending so much time imagining what our technology will be capable of that we’re underestimating the systems we’re relying on to support it?
This (and some other topics) is what we discussed on CJAD 800 AM this week.