OpenAI will delay GPT-5.6 after Trump administration request
- lab Anthropic
- lab OpenAI
- location US
- model Fable 5
- model GPT-5.6
- model Mythos 5
- person Sam Altman
- person Trump
OpenAI has delayed the release of its GPT-5.6 AI model at the request of the US government, with CEO Sam Altman announcing a limited preview to a small group of enterprise customers.
The US government has asked OpenAI to stagger the release of GPT-5.6, with approval for customer access to be granted on a case-by-case basis during the preview period, according to The Information[2]. The Information reported that Altman told staff the general release would follow 'a couple of weeks later' if the process went well. The delay comes as the US government has signed an executive order this month to create a voluntary framework for vetting powerful new AI models before they are released, The Guardian[1] reported. OpenAI's decision echoes the launch of Anthropic's Mythos product, which has since been pulled after the US government ordered the company to restrict foreign access due to its powerful cyber-hacking capabilities. The staggered release was requested after conversations between OpenAI and two government agencies: the Office of the National Cyber Director and the Office of Science and Technology Policy. The Information[2] also reported that former Commerce Secretary Howard Lutnick had intervened against even a limited release, calling Altman to demand approvals from other agencies.
model-release
Background sources we checked (10)
- en.wikipedia.org ↗ OpenAI is an American artificial intelligence (AI) research organization headquartered in San Francisco, consisting of OpenAI Group PBC, a for-profit public benefit corporation (PBC), partially controlled by OpenAI Foundation, a nonprofit. OpenAI develops generative AI models, pa…
- en.wikipedia.org ↗ GPT Image is a series of image generation and editing models developed by OpenAI. A text-to-image variant of the GPT family, it uses deep learning methodologies to generate digital images from natural language descriptions or images precisely. As the successor to DALL-E, GPT Imag…
- en.wikipedia.org ↗ Apple Intelligence is a collection of artificial intelligence features developed by Apple Inc. Relying on a combination of on-device and server processing, it was announced on June 10, 2024, at the 2024 Worldwide Developers Conference, as a built-in feature of Apple's iOS 18, iPa…
- en.wikipedia.org ↗ The following is a list of events of the year 2026 in the United States, as well as predicted and scheduled events that have not yet occurred. July 4, 2026, will be the 250th anniversary of the signing of the Declaration of Independence of the United States from the United Kingdo…
- en.wikipedia.org ↗ Microsoft Corporation is an American multinational technology company headquartered in Redmond, Washington. The company became influential in the rise of personal computers through software like Windows and has since expanded into areas such as Internet services, cloud computing,…
- arxiv.org ↗ We present DarkAgents: a multi-agent system that leverages the reasoning and code-generation capabilities of large language models (LLMs), together with deterministic tested human-written code, to build orchestrated pipelines for theoretical astroparticle physics research. While …
- arxiv.org ↗ Indirect prompt injection in tool-use agents is a concrete production threat: LLM agents read from integrations (third-party services such as Gmail, Salesforce, or Jira accessed through tool calls) whose response content the user neither writes nor controls. Existing benchmarks u…
- arxiv.org ↗ Selecting the right electricity market region for a hyperscale AI datacenter requires reasoning across live electricity prices, grid carbon intensity, technology cost trajectories, and causal grid dynamics -- a multi-step, multi-source analytical task that static knowledge benchm…
- arxiv.org ↗ Coding agents often pass per-prompt safety review yet ship exploitable code when their tasks are decomposed into routine engineering tickets. The challenge is structural: existing safety alignment evaluates overt requests in isolation, leaving models blind to malicious end-states…
- arxiv.org ↗ Existing benchmarks of language-model refusal on malicious-coding tasks routinely conflate requests for executable malicious software with requests for harmful security knowledge. This conflation matters because the two request types plausibly trigger distinct refusal pathways in…