The Attacker Never Sleeps, Neither Can Your Testing

A Snyk blog post by Manoj Nair argues that frontier AI has removed the human time-and-cost constraint on attackers, creating an 'attacker that never sleeps' that reasons about applications at machine speed. Citing Anthropic's disclosure of a state-sponsored group that used its models to run 80-90% of a live espionage campaign against roughly thirty targets, the piece frames AI-accelerated code generation as widening a 'trust gap' where whole vulnerability backlogs become weaponizable.