Environmental Restoration

Fatou knelt in soil that her grandmother said had been sand.

The reforestation AI had mapped every meter of this land: soil composition, water table depth, microclimate variations, root network potential. It had selected 140 species of trees, shrubs, and grasses calibrated to interlock into an ecosystem that would, if the models held, transform this stretch of the Sahel into something that could sustain itself.

Fatou was a planter. Technically the drones could do it. They did do it, for the vast stretches where human hands were too slow. But the program employed human planters at the edges, where the work was more delicate, and where—Fatou suspected—the program’s designers understood that people needed to feel the dirt.

She placed the seedling, a jujube tree, into the hole the drone had pre-dug to the exact depth the AI specified. She packed the soil. She watered it from a solar-powered drip line.

Her grandmother, watching from the shade, said: “When I was a girl, we prayed for rain. Now the machines bring it.”

The machines didn’t bring rain. They brought data, and planning, and the ability to coordinate ten thousand planting decisions across a continent in real time. The rain came from the sky, as it always had, but the AI could tell you when, and how much, and which seedlings would survive it.

Fatou stood and looked at the line of green stretching toward the horizon. It was thin and young and fragile. But it was there. And for the first time in three generations, the sand was retreating.