Some of the numbers are striking. 46% of all AI usage is for work, not personal curiosity. In countries like Brazil and India that figure climbs above 50%. Computer programmers sit at 75% task exposure, customer service reps at 70%. Yet 91% of occupations remain below 30% exposure, and 97% of interactions still involve a human in the loop. The average autonomy score is 3.4 out of 5: collaborative, not autonomous.
That's the optimistic read, and there's real substance to it. People are using AI to draft, iterate, learn, and accelerate. Tasks that took hours compress into minutes. New capabilities unlock for workers who never had access to specialised tools before.
But the data also tells a harder story. The roles with the highest exposure — data entry at 67%, medical transcription at 64%, customer service at 70% — are disproportionately lower wage positions. The benefits of augmentation flow upward while displacement risk concentrates at the bottom. And that 3% of fully autonomous AI interactions today will not stay at 3% for long.
Whether you're a policy maker thinking about workforce transition, a business owner navigating company culture through this shift, or an individual finding your own path, we hope this explorer gives you something concrete to work with. The data is real. The trends are accelerating. How we respond is still up to us.
