Updated
Updated · MIT Technology Review · Oct 28
Only 16% of Organizations Extract Transformative AI Value as Missing Context Layer Blunts Returns
Updated
Updated · MIT Technology Review · Oct 28

Only 16% of Organizations Extract Transformative AI Value as Missing Context Layer Blunts Returns

3 articles · Updated · MIT Technology Review · Oct 28

Summary

  • An MIT Technology Review Insights poll found just 16% of organizations are getting transformative value from AI, while 40% report only moderate gains in specific tasks or teams.
  • The report points to a missing “context layer” between business systems and AI tools, leaving models unable to understand company-specific data, rules, and workflows well enough to act reliably.
  • 38% of respondents are still piloting AI use cases and 35% mainly use co-pilots or chat assistants; only 12% have AI agents in production and 3% use AI for most work in key processes.
  • That gap suggests many companies have adopted multiple AI tools but not the architectural layer needed to turn enterprise knowledge into governed, machine-usable context at scale.

Insights

If every company uses AI, why do only 16% see transformative value—and is a governed context layer the missing piece?
Are enterprise AI agents failing because models are weak, or because they lack trusted business meaning, permissions, and runtime guardrails?
Could the next AI moat be a system of context that knows your company better than your smartest employees?