Meta CEO Mark Zuckerberg has admitted that the company’s drive to re-imagine its workforce around artificial intelligence (AI) has not been flawless. The rapid growth of AI and the scale of organisational change that has taken place has taken time and led to mistakes along the way, he said in an internal memo seen by Reuters.

“As such, we’ve made mistakes and will almost certainly make more,” Zuckerberg said, and Meta is committed to providing as much stability as possible for employees in this transition. His remarks mirror the trade-off between AI innovation and employee morale.

Meta has been investing billions of dollars in AI infrastructure and development, reorganizing teams to fit its long-term strategy. AI is a new field of competition that is only getting hotter as rivals race to be at the forefront of the next wave of technological transformation.

Zuckerberg tried to reassure employees that Meta won’t make any more layoffs in the company this year, which should be a relief to employees after several rounds of restructuring in recent years. But he warned that because the technology is so rapidly changing, it’s hard to predict how it’ll go. “I don’t want to overpromise because the world is changing in ways that are out of our control,” he said.

The acknowledgment of mistakes shows the difficulty large tech companies are facing as they adopt AI-based business models. Meta has made great progress in incorporating AI into its platforms and services, but we still need to know that the human aspects of the transformations are very important. Zuckerberg’s honesty could help build trust in employees and the company is still going through uncertainty.

Finally, Meta’s AI workforce transformation represents the prospect of opportunity as well as risk. Zuckerberg’s assurance of no layoffs in 2026 may provide a little relief for the short run, but they’re an example of how a company can never have a perfect balance between creativity and employee stability in an ever-changing technology.