Yahoo Finance

Trump Administration Asks OpenAI to Stagger Release of AI Model

CNBC

OpenAI limits new AI models to 'trusted partners' at request of U.S. government

R street

The United States Must Reject Government Control of Artificial Intelligence

Assessing Trumps Executive Order on AI Oversight

The Hill

Trump’s last-minute AI order switch exposes White House divides

AdamThierer

arXiv.org

[2307.03718] Frontier AI Regulation: Managing Emerging Risks to Public Safety

MS Now

Trumps executive order on AI is insidious

TechCrunch

The White House is asking OpenAI to slow roll the release of its new model over safety concerns

The Associated Press

OpenAI limits its latest ChatGPT product to Trump-approved customers during cybersecurity review

Guardian

OpenAI staggers AI model release after Trump administration request

Axios

Trump administration asks OpenAI to limit next model release over security concerns

Financial Times

Trump administration asks OpenAI to stagger release of new model to vet users

White House asks OpenAI to limit its next model release

The Washington Post

U.S. government will decide who gets to use latest upgrade to ChatGPT

Politico

Trump administration steps in to limit OpenAI’s latest model launch

Slashdot

Trump Administration Asks OpenAI To Stagger Release of New Model

Bloomberglaw

Trump Administration Asks Openai To Stagger Release Of Ai Model

Trump administration asks OpenAI to stagger release of new model, The Information reports

Metaculus

Cybersecurity Laws for Powerful AI by 2030?

OpenAI announced Friday that its newest AI model, GPT-5.6 Sol, would be released only to a small group of government-approved partners, following a request from the Trump administration to limit the rollout amid concerns over the model's advanced cybersecurity capabilities.

OpenAI announced Friday that its newest AI model, GPT-5.6 Sol, would be released only to a small group of government-approved partners, following a request from the Trump administration to limit the rollout amid concerns over the model's advanced cybersecurity capabilities.

The staggered release was requested by the White House's Office of the National Cyber Director and Office of Science and Technology Policy. Commerce Secretary Howard Lutnick also spoke with OpenAI CEO Sam Altman on Wednesday, seeking assurances that all relevant agencies had tested and approved the model.

The staggered release was requested by the White House's Office of the National Cyber Director and Office of Science and Technology Policy. Commerce Secretary Howard Lutnick also spoke with OpenAI CEO Sam Altman on Wednesday, seeking assurances that all relevant agencies had tested and approved the model.

OpenAI said Sol is its most capable model to date, with improvements in coding, biology and cybersecurity. The company said the model is better at helping users identify and fix vulnerabilities than at carrying out attacks, and does not cross its own "critical" cybersecurity risk threshold.

OpenAI said Sol is its most capable model to date, with improvements in coding, biology and cybersecurity. The company said the model is better at helping users identify and fix vulnerabilities than at carrying out attacks, and does not cross its own "critical" cybersecurity risk threshold.

In a memo to staff, Altman said the government is "approving access customer by customer" during the preview period, and that a broader release could follow "a couple of weeks later" if the process goes well.

In a memo to staff, Altman said the government is "approving access customer by customer" during the preview period, and that a broader release could follow "a couple of weeks later" if the process goes well.

The OpenAI restrictions follow the Trump administration's export control directive against Anthropic, which forced the company to pull its Mythos and Fable models from public access. Negotiations over reinstating access to those models remain ongoing.

The OpenAI restrictions follow the Trump administration's export control directive against Anthropic, which forced the company to pull its Mythos and Fable models from public access. Negotiations  over reinstating access to those models remain ongoing.

Trump signed an executive order earlier in June directing AI companies to voluntarily submit advanced models for government review up to 30 days before release, though the framework's specific rules have not yet been finalized.

Trump signed an executive order earlier in June directing AI companies to voluntarily submit advanced models for government review up to 30 days before release, though the framework's specific rules have not yet been finalized.

Yahoo Finance #

CNBC #

R street #

CFR #

The Hill #

arXiv.org #

MS Now #

TechCrunch #

The Associated Press #

Guardian #

Axios #

Financial Times #

CNN #

The Washington Post #

Politico #

Slashdot #

Bloomberglaw #

Metaculus #

Trump's AI executive order is a giveaway to Silicon Valley dressed up as regulation. The 30-day voluntary review window was watered down from 90 days after Elon Musk, Mark Zuckerberg and David Sacks personally lobbied Trump — and Sacks has financial stakes in hundreds of AI-linked companies. OpenAI itself reviewed an earlier draft, making this less a check on industry power and more a rubber stamp for it.

Trump's AI executive order is a giveaway to Silicon Valley dressed up as regulation. The 30-day voluntary review window was watered down from 90 days after Elon Musk, Mark Zuckerberg and David Sacks personally lobbied Trump — and Sacks has financial stakes in hundreds of AI-linked companies. OpenAI itself reviewed an earlier draft, making this less a check on industry power and more a rubber stamp for it.

Powerful AI models with advanced hacking capabilities demand serious government oversight before public release. Anthropic's Fable 5 had to be pulled after its guardrails proved weaker than advertised, and GPT-5.6 Sol is OpenAI's most capable cybersecurity model yet — real risks that justify a controlled rollout. Pre-deployment risk assessments and government review aren't overreach; they're the responsible baseline for frontier AI.

Powerful AI models with advanced hacking capabilities demand serious government oversight before public release. Anthropic's Fable 5 had to be pulled after its guardrails proved weaker than advertised, and GPT-5.6 Sol is OpenAI's most capable cybersecurity model yet — real risks that justify a controlled rollout. Pre-deployment risk assessments and government review aren't overreach; they're the responsible baseline for frontier AI.

AI: Warfare

AI: Geopolitics

AI: Safety and Regulations

Artificial Intelligence

OpenAI Limits GPT-5.6 Sol Release at Trump's Request

Politics

United States of America

Letting Silicon Valley insiders whisper in the president's ear to kill AI oversight is a dangerous way to run national security policy. Centralizing AI control under the NSA while blocking even voluntary safety reviews moves America backward, not forward. Politicizing which companies get scrutinized and which don't undermines the very innovation edge the administration claims to be protecting.

Letting Silicon Valley insiders whisper in the president's ear to kill AI oversight is a dangerous way to run national security policy. Centralizing AI control under the NSA while blocking even voluntary safety reviews moves America backward, not forward. Politicizing which companies get scrutinized and which don't undermines the very innovation edge the administration claims to be protecting.

Adam Thierer

Powerful frontier AI models can autonomously exploit real-world software vulnerabilities, making pre-deployment government review a genuine national security necessity. A voluntary 30-day review window isn't a regulatory overreach — it's the minimum needed to give defenders a fighting chance before adversaries weaponize the same capabilities. Without structured oversight, the U.S. risks shipping dangerous tools faster than it can defend against them.

Powerful frontier AI models can autonomously exploit real-world software vulnerabilities, making pre-deployment government review a genuine national security necessity. A voluntary 30-day review window isn't a regulatory overreach — it's the minimum needed to give defenders a fighting chance before adversaries weaponize the same capabilities. Without structured oversight, the U.S. risks shipping dangerous tools faster than it can defend against them.

OpenAI Limits GPT-5.6 Sol Release at Trump's Request

OpenAI Limits GPT-5.6 Sol Release at Trump's Request

The Spin

Go Deeper

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters

OpenAI Limits GPT-5.6 Sol Release at Trump's Request

OpenAI Limits GPT-5.6 Sol Release at Trump's Request

The Spin

Go Deeper

Sign Up for Our Free NewslettersSign Up for Our Free Newsletters

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters