{
  "id": 19166093,
  "title": "AI’s new business model: Break the internet, then sell the fix",
  "url": "https://urgent.news/2026/11/09/ais-new-business-model-break-the-internet-then-sell-the-fix",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-09T14:02:49.000Z",
  "source": {
    "name": "Fast Company",
    "slug": "fast-company",
    "url": "https://www.fastcompany.com/91637523/ais-new-business-model-break-the-internet-then-sell-the-fix"
  },
  "original_language": "en",
  "account": "Cybersecurity professionals and digital infrastructure experts are finding themselves in a new reality, one that is both exciting and demanding. Recent events have highlighted the vulnerability of their work, as major hacks have struck Binance, Anthropic, and Amadeus, among others. At the heart of these attacks are generative AI companies like Anthropic and OpenAI, who find themselves caught between empowering hackers and positioning themselves as the only line of defense.\n\nThe severity of the situation is alarming. Since 1988, over 400,000 vulnerabilities have been officially logged, with 13,000 occurring in the last month alone. One particularly dangerous bug, PHP2Shell, affects roughly 70% of websites due to PHP's widespread use. The flaw, which had been known to cybersecurity firms Calif and Wiz since 2024, can enable hackers to gain remote control over a host computer, leading to potential catastrophic consequences.\n\nDevelopers at the PHP Foundation knew about the bug in 2024, but disagreements over a solution resulted in no action being taken. The revelation of the bug has turned a theoretical threat into a flood of attacks. OpenAI and Anthropic, among others, are now emphasizing the danger of these vulnerabilities and the ongoing process of discovering and patching them.\n\nThese AI companies are not only warning about the threat but also offering a solution. OpenAI has launched the Patch the Planet initiative, which aims to identify defects and vulnerabilities in open-source software. Additionally, OpenAI unveiled GPT-6 Cyber, a security-focused model, while Anthropic introduced Claude Mythos Preview. Google has its own cybersecurity model, Gemini Carbon 4.\n\nHowever, these efforts are not without motive. Generative AI companies are desperate for new sources of revenue as overall revenue growth slows. Sam Altman, CEO of OpenAI, even suggested a shift in the defense model, stating that they may have to utilize their own products to create a new kind of defensive shield.\n\nIn essence, these AI companies are exploiting the chaos their products have introduced to monetize the situation. The potential for extortion and the compromise of the open-source commons are significant concerns. While these companies strive to ensure the safety of digital infrastructure, their actions may lead to long-term instability and increased costs for users.",
  "summary": "People who work in cybersecurity or work on the fundamentals of digital infrastructure do not want to live in “exciting times,” but right now, they don’t have a choice. Binance was just hacked in what is, so far, the largest cryptocurrency heist in history. Anthropic, the company behind Claude, narrowly averted a major attack through a third-party extension that would have allowed hackers to…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}