OpenAI COO Brad Lightcap delivered an inconvenient truth at India’s AI Impact Summit: despite 800 million weekly ChatGPT users and $20 billion in revenue, artificial intelligence hasn’t penetrated enterprise business processes.

“We have not really seen enterprise AI penetrate enterprise business process,” Lightcap told attendees, highlighting what the company calls “capability overhang”, which is the gap between what AI can do and what organisations actually do with it.

The admission matters because it exposes the fundamental disconnect between AI’s impressive demonstrations and its operational impact. Companies are buying access, experimenting with tools, and generating impressive proof-of-concept results. But they’re not fundamentally changing how work gets done, how decisions get made, or how value gets created.

OpenAI launched Frontier in January specifically to address this problem, a platform for enterprises to build and manage AI agents that can actually execute business processes rather than just answer questions. The product recognises that individual productivity tools, no matter how sophisticated, don’t transform organisational operations.

“You’ve got really powerful AI systems that any person can use in their individual capacity,” Lightcap explained. “And enterprises are these highly complex organisations with a lot of people, teams, all having to work together, a lot of context. There are very complex goals that have to be achieved using a lot of different systems and tools.”

Consumer AI and enterprise AI solve completely different problems. ChatGPT works brilliantly for drafting emails or explaining concepts. It doesn’t work for automating procurement workflows, managing compliance across jurisdictions, or coordinating cross-functional projects involving dozens of people and legacy systems.

The irony is stark: OpenAI itself remains a massive Slack user, according to Lightcap, demonstrating that even the company building AI still depends on traditional enterprise software for actual collaboration and workflow management.

For Nigerian businesses evaluating AI investments, this admission provides a valuable perspective. The pressure to adopt AI is intense. Every vendor promises transformation. Every competitor claims they’re leveraging artificial intelligence. But OpenAI’s own assessment suggests most organisations are still figuring out what productive AI usage actually looks like beyond individual task automation.

The capability overhang creates strategic implications. First, it means early AI adoption doesn’t necessarily create a competitive advantage if nobody’s actually integrating the technology into core operations. Second, it suggests the real value will accrue to organisations that solve integration challenges rather than those that simply buy access to models. Third, it indicates that AI transformation requires organisational change management, not just technology procurement.

OpenAI’s revenue growth, from $6 billion in 2024 to more than $20 billion in 2025, demonstrates that companies are willing to pay for AI capabilities. But willingness to pay for access differs fundamentally from successful operational integration. Businesses are buying potential, not realised value.

The company plans to measure Frontier’s success based on “business outcomes, not on seat licences,” acknowledging that traditional software metrics don’t capture whether AI is actually changing how work happens. This shift in measurement philosophy reflects recognition that AI’s value lies in operational impact rather than user counts.

For marketers specifically, the capability overhang explains why AI-generated content hasn’t displaced creative agencies, why automation hasn’t eliminated campaign management roles, and why sophisticated language models haven’t fundamentally changed how brands communicate. The tools exist. The integration doesn’t.

OpenAI’s enterprise market share has declined from roughly 50 per cent in 2023 to 27 per cent in 2026, according to Menlo Ventures, suggesting that competitors are capturing growth even as overall adoption increases. This fragmentation indicates that no single platform has solved the enterprise integration challenge decisively.

The path forward requires bridging capability and implementation. Nigerian businesses shouldn’t avoid AI, but they should approach adoption with a clear understanding that buying access to powerful models represents the beginning of transformation, not its completion. The hard work, integrating AI into actual business processes, managing organisational change, and measuring real outcomes, remains largely unfinished across the global economy.

OpenAI’s admission confirms what many suspected: AI’s most impressive capabilities remain largely theoretical for most organisations until someone solves the messy problem of making them work inside complex human systems.

ALSO WATCH:MARKETING EDGE ONTV