Paysend, a London-based payments technology company, is restricting artificial intelligence tool access to its highest-performing employees—a contrarian move in a market racing toward scale.
CEO Ben Chisell, who previously worked at Amazon, eBay and Starling Bank, estimates the company can deliver roughly 25 percent more engineering output with the same team size by using AI selectively. But he enforces a hard discipline: new hires get one month to experiment with AI tools and must present concrete work product before earning permanent access. Chisell said AI can magnify poor judgment and laziness in the wrong hands.
The approach reflects skepticism about AI as a universal productivity tool. At Paysend, when a new search engine optimization lead requested Claude access, the employee used it primarily for producing briefs faster. That narrow use case led to building a dedicated content-brief agent instead, replacing the need for general-purpose access.
Other major UK firms are taking the opposite tack. Lloyds Banking Group rolled out 40,000 Microsoft 365 Copilot licenses with 97 percent active usage. Grant Thornton UK encouraged all staff to use Anthropic's Claude as part of a £500 million investment in AI. Sage, the Newcastle-based software company, said its AI coding tools saved hundreds of thousands of work hours.
Yet adoption has not reliably moved revenue. UK government research from January found three-quarters of businesses using AI reported improved workforce productivity. But 77 percent of those same businesses said revenue had not changed.
Lokdeep Singh, CEO of enterprise software company Unily, shares Chisell's caution. Singh views AI as a "superpower" best reserved for top performers and warned against distributing it broadly as a cost-cutting measure. "I personally believe you should use it to expand your ambition rather than make the team smaller," Singh said.
