AI Disruption

AI Disruption

Share this post

AI Disruption
AI Disruption
Llama 4 Trillion Titan Delayed, 80% of Core Veterans Mass Resign?
Copy link
Facebook
Email
Notes
More

Llama 4 Trillion Titan Delayed, 80% of Core Veterans Mass Resign?

Meta's Llama 4 Delayed, 80% Core Team Quits? AI Race Heats Up Amid Meta's Struggles.

Meng Li's avatar
Meng Li
May 18, 2025
∙ Paid
2

Share this post

AI Disruption
AI Disruption
Llama 4 Trillion Titan Delayed, 80% of Core Veterans Mass Resign?
Copy link
Facebook
Email
Notes
More
3
Share

"AI Disruption" Publication 6400 Subscriptions 20% Discount Offer Link.


80% of the Llama 4 Team Resigns Collectively?

Yesterday, a shocking revelation from a machine learning researcher at AI star startup Prime Intellect set the AI community ablaze.

Meta’s flagship AI model, “Behemoth,” has repeatedly delayed its release, with internal doubts about the company’s multi-billion-dollar AI strategy reaching a peak.

Of the original 14 members of the Llama 1 team, 11 have already left Meta.

In response, Meta executive Dave Arnold quickly debunked the claim of an 80% employee exodus, calling it inaccurate. He emphasized that those who left were only from the Llama 1 paper’s team, not subsequent teams.

Meanwhile, OpenAI launched its powerful coding agent Codex yesterday. Next week, Google’s I/O conference will unveil a series of hits like Veo and Flow.

This turmoil has prompted the industry to ask a soul-searching question: Has Meta hit a bottleneck in this AI race?

This post is for paid subscribers

Already a paid subscriber? Sign in
© 2025 Meng Li
Privacy ∙ Terms ∙ Collection notice
Start writingGet the app
Substack is the home for great culture

Share

Copy link
Facebook
Email
Notes
More