Layoffs sweep through Silicon Valley, frontline engineers reveal the truth: AI efficiency is severely overestimated.

In 2026, there were continuous waves of layoffs in Silicon Valley's tech industry. In January, Amazon confirmed it would lay off about 16,000 people; in February, the fintech company Block laid off nearly half of its employees; in March, Meta was reported to be planning to lay off 16,000 people. The anxiety of white-collar workers being replaced by AI swept through the workplace. However, an article titled "AI Fatigue is Real, But No One is Talking About It" published by software engineer Siddhant Khare from the tech company Ona triggered a widespread discussion among global media and readers. He pointed out the vast gap between the actual application of AI and its ideal vision. He believed that the efficiency improvement brought by AI has been overestimated, and yet employees are experiencing "AI fatigue."