Meta founder Mark Zuckerberg has made a bold prediction that by 2025, artificial intelligence would advance to a level where it can code as effectively as mid-level engineers.

In a recent podcast interview, Zuckerberg revealed plans to replace AI agents with mid-level engineers at Meta labs within the year.

Moreover, at a tech conference, he expanded on his vision, indicating that AI might soon take over many coding tasks currently handled by human engineers.

If AI is capable of writing code with the skill of mid-level engineers, it could lead to significant changes, allowing AI to autonomously manage entire software projects, from initial concept through to deployment.

One speaker humorously proposed the idea of ‘AI Engineers’ Unions’ to advocate for the rights of AI workers, highlighting the significant shift that AI’s integration into engineering could provoke. Although this was a lighthearted comment, it underscored the increasing dialogue about AI’s role in the workforce.

Zuckerberg’s prediction evoked both excitement and concern. On one side, AI-driven software development could enable unprecedented speed and scale in technological innovation. It also raises critical issues regarding job security in the tech industry, the ethical implications of AI usage, and the potential erosion of human decision-making authority in important processes.