Enterprise Deployments Drive Consumer AI Loyalty

PYMNTS Intelligence's Consumer AI Benchmark report, the eighth installment of the series based on an April 2026 survey of 2,663 U.S. adults (including a subgroup of 648 workers whose employer provides or mandates an AI platform), found that 78% of those workers use that same tool for personal tasks: 62% during work hours and 44% after hours, with only 22% keeping it confined to the job. Separately, PYMNTS Intelligence estimates 86 million Americans, 33% of U.S. adults and 53% of the employed workforce, now use AI on the job. The report frames this as a reversal of prior technology cycles: PCs, smartphones and social apps typically reached consumers before employers, while AI is following the opposite path, entering through enterprise procurement and workplace mandates first.

A simple work-to-home crossover rate rewards whichever platform already has the largest consumer base: 91% of ChatGPT's work users also use it personally, versus 77% for Gemini, 68% for Meta AI, 63% for Copilot, and 60% each for Claude and Perplexity. PYMNTS Intelligence argues that ranking is misleading, so it also calculates a "crossover lift": the gap between a platform's personal-use rate among workplace-exposed users and its rate among people who only ever used it outside work. On that measure, Copilot leads at a 10-point lift, Gemini, Claude and Perplexity cluster at 9 points each, and ChatGPT and Meta AI trail at 4 and 1 point, because their personal pull was already strong without a work assist. Applied to PYMNTS' estimated 70 million "Always-On" consumers (27% of U.S. adults, up from 22% in a prior wave), Copilot's lift accounts for about 7.1 million personal users it likely would not have reached without workplace mandates, versus 6.5 million for Gemini, 6.3 million for Claude, and 6.2 million for Perplexity.

PYMNTS Intelligence's own framing cautions against reading Copilot's crossover volume as brand preference: its 63% crossover climbs to 79% among baby boomers and falls to 48% among workers who say AI isn't necessary to their jobs, a split the report calls a distribution signal rather than a preference one. Asked which tool they'd choose without an employer default, consumers picked ChatGPT decisively (named by roughly 83% of iPhone users), and it also topped workplace usage at 73% versus 55% for Gemini and 47% for Copilot. Claude didn't lead on volume, but its work users reported the highest self-assessed dependence, 81% said they couldn't do their job or would be substantially slower without it, a 15-point gap over the 66% average for other platforms, the widest PYMNTS measured. For product and go-to-market teams, the practical takeaway is that bundling, defaults and licensing terms are becoming competitive levers alongside model quality, since they determine which tool a worker is handed in the first place.

PYMNTS Intelligence itself describes the productivity and dependence findings as correlational, not causal, and the underlying numbers come from one proprietary survey series that has not yet been independently replicated by another research firm. Worth tracking: whether vendors start explicitly bundling consumer accounts into enterprise deals, and whether future waves of the Consumer AI Benchmark show the gap widening or narrowing between platforms that are "sticky by default" (like Copilot) and those "sticky by merit" (like Claude and Perplexity) as switching costs change.