should everybody company adopt AI, I would generally agree.
for new AI research to succeed, the three pillars are algorithms, training data, and computation resources. Results are not guaranteed, or may take several years to bear fruit.
But should every company put resources in developing new Algorithms, it is a specialised field and skill is in very high demand so hiring right kind of researchers is very expensive. much of this research is emerging from top academinc institutions and lot of investment is being made for infrastructure and computation resources.
remember we have had atleast 4 AI winters, only last couple of years we have some good success and industry adoption has just started.