MIT on GenAI Divide

$30–40 billion invested… 95% zero return. That’s the striking conclusion of a new MIT report on enterprise GenAI adoption 👉 https://lnkd.in/ep7j_duD

Despite massive investment, the vast majority of organizations are seeing no measurable business impact. Researchers call this the GenAI Divide: a small minority are extracting millions in value, while most remain stuck in pilot mode.

🔎 What’s behind this divide?
It’s not a regulation.
It’s not model quality.
It’s approach.

Key insights from 300 implementations:
📌 Tools ≠ Transformation → ChatGPT, Copilot, and similar tools boost individual productivity, but rarely move the P&L.
📌Enterprise paradox → Big firms launch the most pilots but struggle to scale them.
📌Investment bias → Money flows to “visible” projects (front office), while high-ROI back-office processes are underfunded.
📌Partnership premium → External collaborations succeed at twice the rate of internal builds.

🚧 The real bottleneck? Learning.
Most GenAI systems fail because they don’t retain feedback, adapt to context, or improve with use.
✅ The winners?
Organizations that demand process-specific customization and measure outcomes in business terms—not benchmarks. Vendors who integrate into workflows and enable systems to learn are landing multi-million-dollar deployments in months.
👉 The lesson: scaling GenAI is not about having the most significant budget or the best model. It’s about building systems that fit real work, learn over time, and deliver measurable outcomes.

What do you think? Are we heading for a deeper GenAI Divide, or will these lessons spread quickly enough to close the gap?

This article is part of the project “People and Algorithms in Organisations: Competences to Work in the Digital Environment” (DIGIT_People and algorithms), funded by the NAWA – Narodowa Agencja Wymiany Akademickiej (Polish National Agency for Academic Exchange).

#DIGIT_NAWA #AI #ArtificialIntelligence #adoption #GenAIadoption #GenAIDivide #scalling #AIResearch #NAWA