Hello, dear readers!
This is our weekly brief on remarkable AI topics, so you can stay in the loop without doom-scrolling.
Today's focus — the AI labor apocalypse may be running late. According to a new analysis from Citadel Securities, the current AI boom is actually pushing job postings up, not down. Massive spending on infrastructure is creating demand for engineers and developers. So are the techno-pessimists wrong — or is this just the calm before the automation storm?
Also in this week's edition:
MIT finds that AI punches down on non-native English speakers and the less-educated.
At Mobile World Congress in Barcelona, every other gadget is “AI-first”.
The Rumors of the Labor Market's Demise Are (Seemingly) Exaggerated
If you've been following AI discourse lately, you'd think half the workforce is about to be replaced by chatbots.
Citadel Securities isn't so sure. In a recent macro analysis, the firm argues that the first big economic effect of AI may actually be a hiring boom in unexpected places.
Why? Because building the AI economy takes a lot more than clever models. It takes data centers, chips, cooling systems, power infrastructure, and the people who maintain all of that. According to the report, thousands of new data centers are planned worldwide, and the ripple effects are already showing up in job postings — especially for engineers and software developers.

The report was published in response to the recent viral essay The 2028 Global Intelligence Crisis, which argued that AI-driven automation could soon trigger a collapse in demand for human labor. Citadel’s counter-argument is simpler: technological revolutions rarely eliminate work altogether — they tend to shift where it happens.
History offers some precedent. During the early Industrial Revolution, productivity gains indeed displaced large numbers of workers in traditional crafts. But the same jump in efficiency made factory production dramatically cheaper and more scalable, leading to an explosion in industrial employment. Entire new sectors formed around machinery, logistics, and global trade.
If Citadel’s forecast holds, we might see the same process play out over the next few years — mass firings, like the ones recently announced by Jack Dorsey’s Block, but also a wave of new software businesses that would gobble up the newly unemployed and ask for more and more.
In other words: the labor market might not collapse. It might just get very weird for a while.
AI Is Kind of an Asshole to Foreigners and the Poor
AI tools are often sold as the ultimate knowledge equalizer. A new MIT study suggests the reality isn't quite that simple.
Researchers tested several major chatbots using prompts written by people with different levels of English proficiency and education. The pattern that emerged was uncomfortable: users with weaker English or less formal education often got worse answers.
Sometimes the bots refused to respond. Sometimes the explanations were less accurate or less helpful. In a few cases, the tone even turned a bit… condescending. The same question, phrased differently, could produce noticeably different results.

The takeaway isn't that AI is intentionally biased. But it does show how systems trained mostly on fluent English and highly structured information can struggle when the input looks different.
In theory, AI lowers the barrier to knowledge. In practice, it may still work best for the people who already know how to talk to it.
The First to Jump Into AI-First
If there was one theme at this year's Mobile World Congress in Barcelona, it was this: everything is now an AI device.
Laptop and smartphone makers spent the week pitching hardware designed to run AI models directly on-device instead of sending every request to the cloud. Faster responses, better privacy, less dependence on internet connections — that's the idea.

© 2026 GSMA / MWC
Apple's latest MacBook lineup with its M5 chips leaned heavily into on-device AI performance, while other manufacturers showed off laptops and phones built around dedicated neural processing units.
The pitch is pretty clear: as AI becomes a daily tool — for writing, searching, coding, editing images — people will want machines that can run those tasks locally and instantly.
In other words, the AI boom isn't just about bigger models — it's also about selling you a new laptop.
Thanks for reading AIport. Until next Monday — by then, AI will almost certainly have done something weird again.

