The outside dollar test is the right framework. The problem specific to AI is that the outside dollar and the inside dollar are increasingly mixed on the same invoice. Enterprise AI spend flows through cloud contracts with the same hyperscaler that invested in the lab, bundled with credits and compute commitments from the partnership. The clean separation your case needs between a customer buying value and a partner recycling capital may not exist at the granularity the question requires.
One of the best parts of the article, if not the best for me, was the discussion on customer value creation per unit of compute consumed. This fundamentally alters the formula. Many companies are more focused on achieving full-scale integration without considering other critical factors that could lead to future problems, including the one you mentioned.
Could you publish the reading list for the course? Very interesting topics overviewed.
Is it available for purchase in HBR case collection?
The outside dollar test is the right framework. The problem specific to AI is that the outside dollar and the inside dollar are increasingly mixed on the same invoice. Enterprise AI spend flows through cloud contracts with the same hyperscaler that invested in the lab, bundled with credits and compute commitments from the partnership. The clean separation your case needs between a customer buying value and a partner recycling capital may not exist at the granularity the question requires.
Very interesting topic. And very good arguments on both sides of the equation. Would love to understand the learnings to date.
Thank you!
One of the best parts of the article, if not the best for me, was the discussion on customer value creation per unit of compute consumed. This fundamentally alters the formula. Many companies are more focused on achieving full-scale integration without considering other critical factors that could lead to future problems, including the one you mentioned.