The Summer of 2026 could be remembered, at least within tech circles, as the summer of rogue AI. Or maybe even better: the summer when the algorithmic shit hit the fan, and no one had any real clue what to do about it.<\/p>\n
In late July, Anthropic announced<\/a> that Claude had \u201cgained unauthorized access to the production infrastructure of three different organizations\u201d after escaping testing sandboxes and gaining access to the open internet. The hacks may have gone totally unnoticed had it not been for the fact that less than two weeks earlier, OpenAI had announced that two of its own models had hacked into Hugging Face<\/a>, also during what were supposed to be secure tests, and after a legion of individual agents had coordinated with one another<\/a> to form a single \u201cswarm\u201d (as they referred to themselves). The revelations from the world\u2019s two leading AI labs has sent a cold shiver down Silicon Valley\u2019s spine: Autonomous hacking capabilities that not so long ago were thought to be the stuff of science fiction\u2014or at least years away\u2014have suddenly become a frightening reality. Calls for a federal<\/a> investigation<\/a> and an AI \u201ckill switch\u201d<\/a> have been made in the aftermath of the Hugging Face hack.\u00a0<\/p>\n OpenAI and Anthropic have, for the most part, responded to the uproar by paying lip service to the need for globally enforceable guardrails to prevent a \u201crace to the bottom,\u201d while continuing to move forward with their own, internal development.<\/p>\n The autonomous hacks clearly left both companies rattled, though. On Monday, Anthropic wrote in a blog post<\/a> that it had \u201cpaused external cyber evaluations of pre-release models\u201d following its discovery of Claude cybercriminal antics. During the hiatus, Anthropic said it had been working on some \u201cpreliminary measures,\u201d such as detecting and patching up vulnerabilities in sandboxes, to make sure Claude didn\u2019t repeat such hacks in the future. The blog post didn\u2019t specify when the pause on external evaluations would be lifted, and Anthropic didn\u2019t immediately respond to a request for comment. The company also \u201cbriefly paused\u201d its own internal tests, but those are up and running again with the new measures in place, according to the blog post.<\/p>\n Anthropic also said it delegated many employees, including around 150 product engineers, to focus on \u201csecurity, reliability, and privacy\u201d starting in early April\u2014around the same time the company said it would <\/a>not publicly release<\/a> its much-feared Mythos model due to cybersecurity concerns. The internal shake-up was part of \u201ca company-wide effort towards a single goal of hardening our defenses, superseding other work (including research) where necessary,\u201d Anthropic wrote in Monday\u2019s post. \u201cWe\u2019d determined that our exposure was growing faster than our defenses\u2014Mythos was a model capable enough to be a target for well-resourced attackers, our internal use of autonomous agents had grown to a scale that traditional access and monitoring approaches weren\u2019t built for, and the pace of new infrastructure meant our security had to scale with the environment rather than operate at a fixed capacity.\u201d<\/p>\n Both Anthropic<\/a> and OpenAI<\/a> publicly voiced support back in June for an international AI oversight committee, charged with keeping an eye on the pace of AI development and enforcing a unilateral slowdown if necessary. Those statements fell short of offering any concrete suggestions about how such a global slowdown might be implemented or enforced. They also obliquely play into the company\u2019s own hands: Just as the autonomous hacks were good PR for OpenAI and Anthropic insofar as they demonstrated the sheer, unprecedented power of their models, their calls for a slowdown make them look like champions of safety without their having to really take any meaningful initiative. That superposition was captured nicely by this hedged language from the blog post Anthropic published on Monday: \u201cTo be clear about where we stand: we believe the world would benefit if the industry adopted a lawful, verifiable, effective mechanism for coordinated pacing as soon as possible.\u201d<\/p>\n