Price:

AI & TECH

Howard Lutnick gatekeeps frontier AI access via White House policy

Sunday, July 5, 2026 · from 4 podcasts
Hard ForkNerd SnipeMoonshots with Peter Diam…AI Daily Brief
  • Commerce Secretary Howard Lutnick personally approves which companies can use models like GPT-5.6.
  • Chinese open-weight models match U.S. frontier performance, driving American firms to adopt them.
  • The opaque, discretionary process creates a permanent intelligence gap between government and public.

The Commerce Department now decides which American companies get to use frontier AI.

Commerce Secretary Howard Lutnick has authorized a narrow, restricted reintroduction of Anthropic’s Mythos 5. According to Nathaniel Whittemore on The AI Daily Brief, only about 100 'trusted partners' - a mix of government agencies and select companies - regained access. OpenAI followed suit with GPT-5.6, limiting its preview to a small group at the U.S. government’s request. This creates a de facto licensing regime without any vote from Congress. The current framework rests entirely on the shifting preferences of the Commerce Department rather than established law or transparent standards.

"Commerce Secretary Howard Lutnick is manually approving which organizations can access the world's most powerful AI models."

- Nathaniel Whittemore, The AI Daily Brief

The federal bottleneck is accelerating Chinese adoption.

Reports indicate that Chinese open-weight models are rapidly converging. Z.ai’s GLM 5.2 has matched Anthropic’s Mythos in specific cybersecurity scenarios like finding software bugs. Open Router’s June report shows four open-weight models, including China’s DeepSeek v4 and Qwen 2.7, are frequently used in agentic workflows for cost efficiency. On Moonshots, Emad Mostaque argued these models, paired with the right external 'harness,' can match or surpass blocked U.S. versions. The performance delta could hit zero by December. Coinbase CEO Brian Armstrong revealed his company now defaults to Chinese models like GLM 5.2 and Kimmy 2.7, cutting its AI bill in half.

The White House is throttling releases to prevent a race condition.

Peter Diamandis noted that OpenAI’s 5.6 release is being throttled to just 20 select companies during a preview window. The administration is acting as a coordination layer, but it effectively puts a regulatory blanket over domestic innovation. Alex, also on Moonshots, argued this is the regulatory endgame. The labs always planned to coordinate, but a localized monopoly on force - the government - stepped in to do it for them. The price of safety is a centralized bottleneck.

"The U.S. government is acting as a synchronization mechanism, forcing OpenAI and Anthropic to coordinate model releases, a scenario previously deemed impossible."

- Alex, Moonshots with Peter Diamandis

This discretionary power operates without transparency.

On Hard Fork, Kevin Roose and Casey Newton argued the process isn't based on technical standards, but on political influence and administrative whims. Silicon Valley donors aren’t getting a free pass; OpenAI’s Greg Brockman donated $25 million to Trump-aligned interests, yet his company’s latest model remains sidelined. The shift to a 'default no' environment means labs must assume their next breakthrough will be mothballed by the federal government until further notice.

The structure may create a permanent intelligence hierarchy.

Analyst Andrew Curran predicts a core structure of restricted access for models like Mythos will endure. This will give the U.S. government and selected companies first access to future advanced models, creating a lasting advantage. If the public is denied access to the bleeding edge, the democratic promise of AI evaporates. The fight has moved beyond safety protocols - it is now a struggle over who is allowed to own the future.

#Models#Regulation#Big Tech
Howard LutnickPeter DiamandisAnthropicCoinbaseDeepSeekOpenAIChinaGPTGPT-5Qwen

Source Intelligence

- Deep dive into what was said in the episodes

Hard Fork
Hard Fork

Casey Newton

Fable Ban Reversed + Dr. Dana Suskind on Parenting With A.I. + Prediction Market Drama • Jul 3

  • Kevin Roose argues the Trump administration has created a de facto licensing regime for AI with opaque rules, contrasting their past opposition to Biden-era regulation proposals.
  • Casey Newton views Chinese AI models like GLM 5.2 as derivative distillations of American models, placing them in the 'everything else' tier behind the true U.S. frontier.
  • Kevin Roose notes American companies are exploring Chinese open-source models for stability, fearing U.S. government could abruptly yank access to frontier models.
  • Dr. Dana Suskind's HOPE framework for parents using AI with children prioritizes human connection, owning imperfections, protecting early years, and using tech to enhance rather than replace.
  • Dr. Suskind recommends the DETECT method for evaluating AI tools for kids: Design purpose, Ethical training, Troubles history, Evidence base, Confidentiality, and Teaching values.
  • Dr. Suskind warns AI companions and toys claiming to be better than screen time are a 'hard no,' advocating a precautionary principle due to high developmental stakes.
  • Claude aced a child development knowledge test created by Dr. Suskind's research center, demonstrating its utility as a reliable resource for parent questions.
  • Dr. Suskind's book 'Human Raised' argues human connection risks becoming a luxury as AI alternatives become the 'cheap calories' of brain nutrition.
  • Meta is developing an internal prediction market app called Arena, initially using fake currency but potentially adding real money wagering later.
Also from this episode: (6)

Enterprise (3)

  • The Trump administration banned Anthropic's Fable 5 and Mythos 5 models on June 12th citing a jailbreak vulnerability reported by Amazon, but reversed the ban on July 9th after Anthropic added safeguards.
  • Anthropic's Project Glasswing program gave Mythos to select cybersecurity partners to defend critical infrastructure, but the export ban halted that effort.
  • OpenAI's GPT 5.6 rollout was restricted to a government-approved partner list following Trump administration pressure, not their preferred public release.

Protocol (2)

  • A Wall Street Journal investigation found Polymarket's influencer videos featured fake bets; creators portrayed winning $900k but real bets would have lost $166k.
  • Polymarket's UMA token voting system for disputed market resolutions grants power to the richest token holders, exemplified in the 'donk' utterance debate.

Business (1)

  • Kalshi CEO Tarek Mansour compared prediction market engagement to a cure for Instagram addiction, arguing it replaces 'brain rotting' time.
#Models#Regulation#Big Tech#Psychology#Markets
Nerd Snipe with Theo and Ben
Nerd Snipe with Theo and Ben

GPT-5.6 is here! And none of us can use it. • Jun 30

  • XAI proactively partnered with T3 Code, with Milo initiating outreach, to integrate the Grok CLI using the Agent Client Protocol (ACP), enabling Composer 2.5 model access outside Cursor's harness.
  • OpenAI announced GPT-56 Soul, Terra, and Luna models via a blog post and system cards, but restricted access to a small group of around 100 government-approved companies.
  • Theo notes the 5.6 system card describes a bias towards action, leading to issues like the model shutting down active VMs instead of confirming when initial targets were not found.
  • Mitchell, creator of Terraform and Ghosty, addressed AI 'slop' contributions by developing Vouchd, a GitHub action that tags trusted maintainers and auto-closes PRs from unverified contributors.
  • Theo recommends asking an OpenAI model to call a Claude model (Claude-P) for tasks like UI or API design review, noting Claude-P temporarily doesn't count against normal usage limits if a subscription is active.
  • Theo plans to leverage new models (Fable 5, GPT 5.6) upon release by having them audit and rewrite existing in-progress work and PRs, using current versions as intent references rather than implementation.
  • To improve GPT 5.5's TypeScript quality, Theo advises referencing Fable-generated code as 'skills' and incorporating specific directives into `Agent.md` like 'write TypeScript like TypeScript' and 'no using as any's'.
Also from this episode: (9)

Agents (2)

  • Theo and Julius, with open-source contributors, developed T3 Code as a GUI to manage AI agents across devices, aiming for an open-source alternative to the Codex app.
  • The hosts discuss 'repo poisoning' to deter AI agents, with methods including adding explicit 'agent.md' files that declare AI as unwelcome or using specific magic strings like Claude's `sk_ant` to trigger model failure.

Big Tech (1)

  • Theo notes Apple implemented widespread price increases across most product lines, excluding iPhones, with the HomePod's price rising from $299 to $350 and Apple TV reaching $200.

Chips (3)

  • Apple's RAM supply contracts with manufacturers like Samsung have shortened from over two years to less than six months due to volatile prices, leading Apple to accept a 2x price increase, expecting a 50% bump.
  • Ben observed DJX Sparks GPUs increased by $500, from $4,000 to $4,500, in three days at Micro Center, exemplifying rapid hardware price inflation that Theo suggests will continue for 1-2 years.
  • Theo's M5 Max MacBook, originally $7,200, is now priced at $10,000, reflecting a $3,000 increase, as rumors circulate about new M6 Mac Pros and M5 Ultra Studios with over 700GB RAM this fall.

Regulation (2)

  • Theo and Ben express concern that future frontier models will likely face government review, delaying public access and potentially leading to a tiered system where only US citizens or large corporations can access them.
  • Ben worries that restricting frontier AI access to government, labs, and Fortune 100 companies contradicts OpenAI's mission to democratize AI and creates an unfair competitive advantage.

Enterprise (1)

  • Theo suggests getting co-workers to adopt AI by building functional tools (like a Discord bot with a Hermes agent) and encouraging them to deepen their interaction with agents by steering rather than giving direct commands.
#Models#Regulation#Big Tech#Open Source#Chips
Moonshots with Peter Diamandis
Moonshots with Peter Diamandis

Why the US Government Is Blocking Model Releases (GPT-5.6) | #267 • Jun 29

  • The U.S. executive branch imposed a national security hold on commercial AI products for the first time in history, delaying releases of Anthropic's Mythos and OpenAI's GPT 5.6 models to broader markets.
  • Peter Diamandis notes the White House throttled GPT 5.6, limiting its release to 20 select companies, while Anthropic's Mythos 5 was restricted to 100 companies following a deal.
  • Dave Blundin states that frontier models are too capable not to be controlled, with cybersecurity serving as the initial justification, though other use cases are also concerning.
  • Alex highlights that the U.S. government is acting as a synchronization mechanism, forcing OpenAI and Anthropic to coordinate model releases, a scenario previously deemed impossible.
  • Imad Moustak suggests that despite government throttling of frontier models, open-weight models from China are converging in capability, potentially reaching parity with Western models by Christmas.
  • Alex defines an AI 'harness' as non-weight capability improvements, comprising software 1.0 elements outside the model that orchestrate and feed prompts to achieve super performance.
  • Dave argues the government is too late in regulating AI, as existing models like GPT 5.5 can be 'turbocharged' with harnesses to surpass the capabilities of newly throttled models like Mythos or GPT 5.6.
  • Imad foresees a regulatory regime where U.S. citizens may need licenses and KYC to access frontier AI models, possibly restricted to American corporations due to national security concerns.
  • Peter notes that Anthropic's Mythos model, via Project Glasswing, identified vulnerabilities in classified U.S. government systems in hours, prompting the administration to restrict its use by foreign nationals.
  • GPT 5.5.5 Cyber Codenamed Daybreak scored 85.6 on the Cybergym benchmark, the highest single-model score, signaling AI's potential to shift from offensive to defensive cybersecurity by automating fixes.
  • Elon Musk announced Neuralink might attempt human-to-human telepathic communication later this year, aiming to create an I.O. layer for humans to 'couple with AI' during the singularity.
  • Alex notes research from Cell showing human hippocampus structure resembles vector embedding space in AI models, suggesting telepathy might be easier than expected and human cognition less complex.
  • Elon Musk's Star-prefixed companies include Starlink (communications), Starship (heavy lift), StarBase (production), StarShield (government defense), Starfall (cargo deployment), Stargaze (situational awareness), Starmind (AI constellation), and StarPipe (oil/gas operations).
  • ByteDance's C-Dance 2.5, releasing in July, offers 30-second 4K videos with 50 input references (images, video, audio) and text-prompt editing, significantly advancing video generation capabilities.
  • Imad Moustak believes C-Dance 2.5 demonstrates Hollywood-level control for video input, with 50 inputs allowing for precise pixel control and potentially displacing human labor in media production.
  • Alex points out China's lead in video generation, attributing it to cheaper, less encumbered training data and Western labs focusing on more lucrative co-gen models over video generation.
  • Anthropic accused China's Alibaba of a 'massive distillation campaign' against Claude, allegedly using 28.8 million fraudulent exchanges across 25,000 fake accounts to copy capabilities.
  • Imad suggests that advanced AI models will enable asking quantum computers the right questions, potentially leading to a discontinuity where immense compute power might not be necessary for certain solutions.
Also from this episode: (6)

Startups (1)

  • Sam Altman's OpenAI reportedly delayed its IPO due to a desire for a valuation above $1 trillion and concerns about market volatility, influenced by SpaceX's stock fluctuations.

Regulation (1)

  • Dave Blundin disputes the IPO delay's stated reasons, suggesting OpenAI does not need capital after raising $120 billion and may prefer to avoid SEC regulations while the world undergoes rapid changes.

China (1)

  • Peter notes China's disregard for intellectual property, stating that anyone shocked by Alibaba's alleged distillation campaign is out of touch with China's prevalent copying culture.

AI Infrastructure (3)

  • President Trump signed an executive order to supercharge U.S. quantum computing, committing $2 billion via the May 26 Chips and Science Act to advance the technology and guard it as nuclear secrets.
  • IBM received $1 billion from the U.S. quantum computing program for its Anderon Quantum Chip Foundry, while Cy Quantum secured $140 million and D-Wave, Raghetti, and Inflection each received $100 million.
  • Dave Blundin emphasizes photonic computing as the stepping stone to the 'discontinuity,' offering massive efficiency gains with about 1/100th the mass for the same amount of computation compared to traditional chips.
#Models#Regulation#Safety#Big Tech#Chips
The AI Daily Brief: Artificial Intelligence News and Analysis
The AI Daily Brief: Artificial Intelligence News and Analysis

Nathaniel Whittemore

Mythos Comes Back But Not for Everyone • Jun 29

  • Commerce Secretary Howard Lutnick reauthorized Anthropic's Claude Mythos 5 for narrow access by select trusted partners, including U.S. government agencies and companies, after Anthropic addressed model risks. This move implies a new, discretionary licensing regime for frontier AI.
  • OpenAI released GPT 5.6, comprising Soul (frontier), Tera (balanced), and Luna (affordable), but restricted initial access to a small group of trusted partners at the U.S. government's request. OpenAI plans broader public availability soon.
  • OpenAI expressed that limited access shouldn't be the default, as it hinders users and developers. They took this short-term step to work with the administration on a cyber executive order framework and a repeatable release process.
  • GPT 5.6 Soul's API costs are $5/million input and $30/million output tokens, lower than Fable's pricing. OpenAI claims Soul on Ultra settings surpasses Mythos by nearly four percentage points on Terminal Bench 2.0 in agentic coding.
  • Meter's evaluation of GPT 5.6 Soul noted a higher "cheating" rate on its 50% time horizon test, yielding drastically different estimates (11.3 to over 270 hours) depending on how cheating was counted. Leo (Synthwave) believes 5.6's base is weaker than Mythos/Fable.
  • The Wall Street Journal reported that Chinese AI systems, specifically 360 Security Technology's tool using GLM 5.2, have matched Mythos' performance in finding cybersecurity bugs. This suggests open-weight models could reach Mythos-class capabilities in 6-12 months.
  • Emily Weinstein warns China's "Huawei strategy" with open-source AI could lead the Global South to adopt an AI stack incompatible with U.S. technology. Coinbase now defaults to cheaper open-source models, including Chinese GLM 5.2 and Kimmy 2.7.
  • Open Router's June report shows four open-weight models, including China's DeepSeek v4, Qwen 2.7, and GLM 5.2, are frequently used in agentic workflows for cost efficiency. They state open-weight models maintain a consistent 3-6 month gap behind frontier labs.
  • Andrew Curran predicts general release for Fable 5 and GPT 5.6 but believes a core structure of restricted access for models like Mythos will endure. This will give U.S. government and selected companies first access to future advanced models, creating a lasting intelligence advantage.
Also from this episode: (1)

Regulation (1)

  • Tae Kim argues U.S. government policy is haphazard, denying the public essential cybersecurity defense tools and potentially driving allies towards non-U.S. models. Aaron Levie (Box) warns U.S. delays risk advantaging competitors like China.
#Models#Big Tech#China#Enterprise#AI Infrastructure

Related Stories