AI Challenges Traditional Technology Pricing

Artificial intelligence is disrupting the link between effort, cost and value that has shaped technology services pricing for decades, according to Amar Aswatha, a senior vice-president of global business engineering at CGI.

Aswatha says AI-enabled operations can deliver results with less manual work. In CGI client engagements, auto-generated tickets have fallen by as much as 70%, about one in five operational events have been resolved autonomously, and predictive models have been associated with up to 45% fewer major outages. The figures are based on the company’s client work, and the source does not provide further detail on how they were measured.

That shift complicates contracts built around staffing, fixed scope and stable demand. Aswatha says organizations need to start with operational problems, redesign processes where appropriate and establish credible baselines before agreeing on the value AI should deliver.

Pricing models each bring trade-offs. Outcome-based fees can be difficult when several automated steps contribute to a result. Consumption-based pricing can become unpredictable as AI agents run in the background. As one example, Aswatha describes a banking client’s 20-developer modernization program, where monthly token costs per developer reached roughly $1,000. Routing simpler work to lower-cost models, identifying unusually high usage and automating repetitive tasks cut that figure to $200.

Aswatha says hybrid pricing may suit AI-enabled managed services: a predictable base fee, a variable charge tied to AI consumption and incentives linked to measurable outcomes. He also calls for clear baselines and scheduled reviews of scope and pricing.