Overview
- Mark Zuckerberg publicly rejected the idea that AI must cause mass job losses and promoted a “personal superintelligence” approach that he says would raise individual productivity rather than simply automate work.
- Meta has doubled down on AI investment and product work, releasing the Muse Spark language model and standing up a SuperIntelligence Lab that Zuckerberg says is under a year old and showing progress.
- The company cut roughly 8,000 roles and closed about 6,000 open positions in May and has reassigned thousands more into AI tasks, moves that have sharply hurt morale and prompted employee protests over surveillance and repetitive training work.
- An internal program that captured employee clicks and keystrokes to train AI agents was paused after staff raised privacy and access concerns, underscoring tensions between Meta’s data collection for model training and worker trust.
- The episode highlights a broader split in tech about whether AI will centralize work or empower individuals, and it leaves regulators, customers and employees watching whether Meta can convert its hiring, spending and new hardware bets—like Meta Glasses—into durable, widely used tools.