Companies rethink rapid AI rollouts as errors and costs rise

Some companies are slowing their rapid AI rollouts after finding that the tools can create errors, low-quality work and extra costs.

Meta has scaled back plans to cut staff and use more AI agents after employees spent increasing time fixing technical and security problems.

At Microsoft, an executive warned that AI-generated writing can produce an “ocean of sameness” rather than new ideas, while Walmart workers have faced errors and false alerts from tools that assign tasks and monitor performance.

Nine in 10 executives said AI had not yet raised productivity at their companies, according to an Atlanta Federal Reserve working paper.

A separate survey found that 36% of senior AI executives at large companies reported an AI incident involving problems such as data loss or operational disruption.

Some firms are restricting AI use, raising security concerns and costs, while also bringing back experienced workers and increasing training.

Microsoft’s AI strategy chief said that giving employees access to tools does not by itself transform a company, and businesses are still working out how to use AI effectively.