Most AI discussions are still about models.
I think the real challenge lies elsewhere.
MIT NANDA found that 95% of AI pilots showed no measurable to P&L impact. The bigger gap was not model capability alone, but getting AI to integrate with and fit into real enterprise workflows.
In other words, AI didn't fail.
The operating system around AI did.
This is also why we are seeing the emergence of a new role, the Forward Deployed Engineer (FDE).
Someone who understands both technology and business, and can bridge the gap between an AI model and a company's day-to-day operations.
The market has recognised the importance of this capability. Demand for FDEs has grown rapidly, with companies like OpenAI, Anthropic, Microsoft and Palantir investing heavily in building such teams.
I don't think the takeaway for a mid-market company is that it now needs to compete with frontier labs for the most expensive AI talent.
The takeaway is that this capability needs to exist somewhere.
→ AI creates value only when it becomes part of business workflows, not when it sits as a standalone pilot. → Models are becoming increasingly accessible. Integrating them into real business processes is where the competitive advantage lies. → Some organisations will build this capability internally. Others will partner with firms that already have it. Both are valid approaches.
At Indus Net Technologies (INT.) this is precisely the capability we have been investing in, not just implementing AI, but helping organisations redesign workflows, data flows and governance so AI can create measurable business value.
Technology is rarely the bottleneck.
Execution almost always is.
Has your organisation invested only in AI tools, or also in the capability to make them work?