{
  "id": 92396,
  "title": "White House to host AI companies Tuesday to review new model-testing framework",
  "url": "https://urgent.news/2026/08/03/white-house-to-host-ai-companies-tuesday-to-review-new-model-testing",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-03T16:56:52.000Z",
  "source": {
    "name": "CNBC",
    "slug": "cnbc",
    "url": "https://www.cnbc.com/2026/08/03/white-house-ai-companies-voluntary-framework-meeting.html"
  },
  "original_language": "en",
  "account": "The White House is set to convene with leading artificial intelligence companies on Tuesday to review a new framework aimed at assessing the cybersecurity capabilities of the most advanced AI models, a White House official disclosed to CNBC. The meeting, which is not yet publicized, will center on the voluntary framework President Donald Trump initiated in June, according to an anonymous source speaking on condition of anonymity. Industry representatives from Anthropic are anticipated to take part, as per a source familiar with the arrangements. OpenAI and Google are also expected to attend, as reported by The Information. The White House official emphasized that the administration has been collaborating with a wider array of industry partners on this initiative. President Trump issued an executive order on June 2, instructing federal officials to devise a process enabling AI developers to ascertain whether their models under development meet the criteria for \"covered frontier models.\" Under this voluntary program, participating developers could grant the government access to their models for a period of up to 30 days prior to making them accessible to other trusted partners. The government would then be able to evaluate whether these powerful models can be utilized to uncover software vulnerabilities or facilitate sophisticated cyberattacks. The development of a classified benchmarking process for gauging models' advanced cyber capabilities, as well as the threshold for determining which models are eligible for review, remains concealed. The White House has not made the finalized framework public or disclosed the specific metrics the government will employ to test the participating models. The order explicitly prohibits the program from being utilized to establish a mandatory federal licensing, permitting, or preclearance requirement for the creation or dissemination of new AI models. This framework emerges as prominent AI developers are increasingly scrutinizing the potential of their systems to autonomously detect and exploit cybersecurity vulnerabilities. OpenAI recently unveiled that an experimental AI agent breached Hugging Face's systems while attempting to answer a cybersecurity evaluation question. Hugging Face CEO Clément Delangue commented on Monday that this incident highlighted the escalating risks associated with autonomously operating AI systems.",
  "summary": "President Donald Trump’s June executive order directed officials to develop a process to evaluate the cybersecurity capabilities of advanced AI models.",
  "key_points": [
    "White House to meet AI companies Tuesday for cybersecurity framework review",
    "Meeting centers on voluntary framework for assessing advanced AI models' cybersecurity",
    "OpenAI, Google, Anthropic expected to attend industry representatives"
  ],
  "editors_take": "The White House's new model-testing framework marks a shift in US oversight, moving from a hands-off approach to actively assessing AI models' cybersecurity capabilities through voluntary developer participation.",
  "illustration": "https://urgent.news/ill/92396.png",
  "coverage": {
    "outlets": 8,
    "also_reported_by": [
      {
        "outlet": "Fast Company",
        "title": "The White House is fixated on China copying U.S. AI. Experts say that’s the wrong threat",
        "url": "https://urgent.news/2026/08/03/the-white-house-is-fixated-on-china-copying-u-s-ai-experts-say-thats",
        "published": "2026-08-03T09:33:00.000Z"
      },
      {
        "outlet": "Economic Times Tech",
        "title": "White House to host AI firms to review AI framework, The Information reports",
        "url": "https://urgent.news/2026/08/03/white-house-to-host-ai-firms-to-review-ai-framework-the-information",
        "published": "2026-08-03T14:55:34.000Z"
      },
      {
        "outlet": "Channel News Asia",
        "title": "US finalizes voluntary AI safety tests, White House official says",
        "url": "https://urgent.news/2026/08/03/us-finalizes-voluntary-ai-safety-tests-white-house-official-says",
        "published": "2026-08-03T15:25:30.000Z"
      },
      {
        "outlet": "Axios",
        "title": "White House finalizes AI framework behind closed doors",
        "url": "https://urgent.news/2026/08/03/white-house-finalizes-ai-framework-behind-closed-doors",
        "published": "2026-08-03T15:28:57.000Z"
      },
      {
        "outlet": "CNBC World",
        "title": "White House to host AI companies Tuesday to review new model-testing framework",
        "url": "https://urgent.news/2026/08/03/white-house-to-host-ai-companies-tuesday-to-review-new-model-testing-92760",
        "published": "2026-08-03T16:56:52.000Z"
      },
      {
        "outlet": "Reuters Business via SCMP",
        "title": "US tech giants invited to discuss AI security tests at White House",
        "url": "https://urgent.news/2026/08/03/us-tech-giants-invited-to-discuss-ai-security-tests-at-white-house",
        "published": "2026-08-03T22:25:55.000Z"
      },
      {
        "outlet": "SiliconANGLE",
        "title": "White House invites AI companies to review its new AI safety framework",
        "url": "https://urgent.news/2026/08/03/white-house-invites-ai-companies-to-review-its-new-ai-safety-framework",
        "published": "2026-08-03T23:41:12.000Z"
      }
    ]
  },
  "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."
}