Both Sides Are Lying (A Little)
The pessimists say: "AI eliminated 55,000 jobs in 2025. Entire professions are disappearing. Your job is next." The optimists say: "AI created 1.3 million new jobs. It's the biggest opportunity in a generation. Stop worrying." Both claims are technically true. Both are misleading. The real story is far more nuanced — and far more useful — than either camp wants to admit.
Let's look at what actually happened.
The Numbers That Scare People
AI directly displaced approximately 55,000 US jobs in 2025. Entry-level hiring dropped 73.4% year-over-year in AI-exposed roles. Workers aged 22-25 in AI-exposed occupations saw a 13% employment decline. By 2030, activities accounting for 30% of US work hours and 27% of European work hours could be automated. The WEF estimates 92 million jobs will be displaced through 2030. 14% of employees globally may need to change careers entirely.
These numbers are real. They represent real people losing real livelihoods. Anyone who dismisses them is either not paying attention or not being honest.
The Numbers That Should Reassure People
LinkedIn data shows AI added 1.3 million new jobs in the same period those 55,000 were lost. The number of workers in AI-fluent roles grew sevenfold in two years — from 1 million to 7 million. There was a 70% year-over-year increase in US roles requiring AI literacy. AI-skilled workers earn a 56% wage premium. The WEF projects 170 million new roles will be created by 2030 — net positive of 78 million jobs even after displacement.
These numbers are also real. The job creation is outpacing displacement by a wide margin. Anyone who ignores them is either not looking at the data or not being fair.
The Honest Take: It's a Transformation, Not a Destruction
The data tells a clear story if you're willing to read it without a predetermined conclusion: AI is causing massive job churn, not massive job loss. Roles are being created faster than they're being destroyed. But the new roles often require different skills than the old ones. And the transition is not evenly distributed.
Who gets hurt: workers in roles that are primarily information processing — data entry, basic analysis, routine content creation, standard bookkeeping. Workers without access to AI training. Workers in companies that automate instead of augment. Workers under 25 who are entering a job market where the first rung of the ladder is being replaced by AI. Workers in lower-wage positions, who are 14 times more likely to need career changes than high-wage workers.
Who benefits: workers who develop AI fluency quickly. Workers in roles that combine AI capabilities with human judgment. Workers in organizations that invest in reskilling. Workers who can articulate the parts of their work that AI genuinely can't do — and there are more of these than most people realize.
The uncomfortable middle ground: most workers will neither lose their jobs nor be unaffected. Instead, their jobs will change. Some tasks will be automated. New tasks will appear. The job title might stay the same, but the actual work will be different. The accountant still calls herself an accountant — but she spends 60% less time on compliance and 60% more time on advisory. The marketer still calls himself a marketer — but content creation is now 20% of his time instead of 80%.
The Inequality Problem Nobody Wants to Discuss
Here's the piece that both optimists and pessimists consistently dodge: AI's benefits and harms are not distributed equally. AI-exposed sectors show 3x higher revenue growth per worker and 4.8x faster productivity growth — for the companies and workers who capture those gains. The 56% wage premium exists — for those who have AI skills. The 1.3 million new jobs exist — for those who can access and qualify for them.
Meanwhile: 20% of ASEAN's population lacks internet access. Over 90% of enterprises face critical AI skills shortages. Only 9% of organizations have achieved AI maturity. The gap between AI "haves" and "have-nots" is widening at every level — between countries, between companies, between individuals.
This means the most important question isn't "will AI take my job?" It's "which side of the transition will I end up on?" — and that question is answered not by headlines, but by what you do in the next 18 months.
Your Move
Stop reading predictions. Start reading job postings. Search LinkedIn for roles in your field that mention "AI." Count them. Compare to six months ago. That's your personal signal — not a headline written by someone who doesn't know your industry, your market, or your skills. The aggregate data matters, but YOUR data matters more. Your AIGMI score gives you a starting point. What you do with it determines which side of the churn you land on.
AI isn't taking all the jobs. It's reshuffling them. The question is whether you're paying attention to the new deck.