New Delhi, Sept. 16 -- Artificial intelligence is putting the traditional headcount-led business model of technology services companies under pressure, as falling AI costs and rising automation reduce the amount of human effort required to deliver software and business processes, according to a new report by Boston Consulting Group (BCG).
The report, Now that (almost) Everyone has AI, says AI is different from earlier technology waves such as cloud and digital transformation because it can directly replace work that technology services companies have traditionally billed for.
The cost of GPT-4-class AI inference has fallen about 60-fold in less than two years, while more than 30 major frontier model versions were released by four leading AI companies between 2023 and mid-2026, BCG said. Work that once required a team of 10 developers for six months can now potentially be prototyped in days.
"AI was supposed to disrupt everyone. Instead, it's starting with the people who built it," Akash Bhatia, managing director and senior partner and global sector leader for technology at BCG, said in a recent post. He said smaller teams and fewer hires were already allowing individuals to do work that previously required many people, adding that industries built around adding people to grow revenue were exposed to the shift.
For IT services companies, the implication is a potential squeeze on the traditional model of selling seats, licences or time and materials. AI agents do not necessarily operate like human employees, making per-seat pricing less relevant, while greater automation can reduce billable hours.
Buyers want business value
BCG's analysis of more than 70 conversations with enterprise leaders found six recurring demands from technology buyers: clear business value tied to profit-and-loss outcomes, solutions grounded in the way work actually operates, integration with existing technology stacks, data sovereignty and governance, in-house control and proof of value with accountability for results.
Yet the report found that technology providers are largely competing on similar technical capabilities. More than 90% of the providers analysed talk about multi-agent orchestration, autonomous task execution and agent architectures.
"Most players are crowded onto the tech story," the BCG authors said, arguing that providers are not talking enough about who owns the business result. The analysis covered more than 40 providers across hyperscalers, SaaS platforms, business process outsourcing companies, systems integrators and AI-native firms.
Karan Chadha, managing director and partner at BCG, has focused on the same shift in the technology services market. His work at BCG examines how AI and agentic AI are reshaping business models, positioning, differentiation and value creation for technology companies.
The report found that fewer than 20% of providers directly address cultural and operating-model changes, even though more than 45% of enterprises identify such changes as a major barrier to AI adoption. More than 70% of providers discuss the importance of data, but fewer than 25% address differentiated data infrastructure, semantic graphs or data layers.
From hours to outcomes
The pressure is particularly visible in outsourced operations such as business process outsourcing, managed services and customer experience, where processes are repetitive, and outcomes are easier to measure.
BCG contrasts the traditional "access and effort" model with an "operator" model, where providers are paid for the work completed or outcome delivered. For every $1 an enterprise spends on software, it spends roughly $6 running it, the report said, making the operational layer a potentially larger opportunity for providers that can combine AI with domain expertise.
The report argues that providers with deep knowledge of a customer's processes, exception patterns and operating environment have an advantage if they can embed that knowledge into AI systems and take responsibility for outcomes.
That shift is already influencing software pricing. Bhatia's earlier BCG work on the agentic AI era argues that seat-based pricing becomes increasingly difficult to sustain when autonomous systems replace the users who traditionally occupied those seats.
For established IT services companies, however, moving to outcome-based contracts is not straightforward. Procurement processes, legal concerns and clients' reluctance to change established commercial arrangements can slow the transition.
BCG recommends starting with one process and one client where a provider has deep domain expertise and the outcome can be measured. The provider can then build a dedicated operating team, embed process knowledge into the AI layer and align its commercial model with the result delivered.
The larger shift, according to the report, is from selling hours to owning outcomes. As AI makes technology execution cheaper and more widely available, domain knowledge, client relationships and the ability to take responsibility for measurable business results could become more important sources of differentiation for technology services companies.
Published by HT Digital Content Services with permission from TechCircle.