Deep learning based body growing

Just my brief sci-fi here:

2031.
Google Health has learned how to grow human organs.
The approach is simple:
A grid in the form of a sphere with a diameter of 20 cm is fixed in a physiological solution, on the surface of which bacteria live.
Bacteria can produce various growth factors, etc., which affect the development of organs in embryogenesis. The expression of these factors is regulated by optogenetics. Lasers shine differently on different parts of the sphere, and different concentrations of factors and their gradients are created.
Stem cells are fixed in the center of the sphere.
A database is being accumulated on the evolution of stem cells depending on the protocol of optogenetic illuminations.
AlphaZero_GrowOrgan predicts the development of an organ and selects the right protocol.
The complete correspondence of growing organ to its human counterpart (say, the heart / kidneys / lungs / gastrointestinal tract) is not strictly necessary. It is enough for it to pump / filter / saturate the blood, be about the right size and shape.
It is difficult to say who was the first to grow the human body according to this scheme, because there were many attempts, and they gradually turned out better and better.
In 2039, the first successful transplant of an old head onto a new body was performed with spinal cord fusion. The patient feels well, the new body is not yet working at full capacity, because it's in compatibility mode with the old brain. Although the patient was vaccinated against Alzheimer's disease in 2034, the condition of the brain vessels needs additional vascular therapy. Stay tuned!

See also Futuristic cell reprogramming.