This is an earlier edition. Today's AI Daily →

DailyAISat, 14 Nov 2026Published 10:14 UTC
Compute, chips and supply: export controls and China's compute gap
OHMEGA / MONSTERAI Daily Sat, 14 Nov 2026AI-generated image — not a photograph

Nvidia CEO Jensen Huang resigned amid probe and Blackwell-in-China claims escalated export-control scrutiny while other AI theaters held in stasis

The day in 30 seconds

All situations side by side · state, direction, Pulses, coverage
State · move in 24hΔ 24h
  1. Compute, chipsStasis since Nov 9 broke rising
    model-assessed
    ◆ 0 rising
  2. Open-weights race: Muse Spark Open, Nemotron, BeamNothing has changed since Nov 11: open-weights remain in gradual catch-up to the closed frontier… steady— rising
  3. Neocloud financing stressNo new Texas moratorium, project halt or financing deal was confirmed… steady
    model-assessed
    ◆ 0 falling
  4. AI safety governance, evaluationPattern since Nov 10 held through Nov 14: steady, gradual escalation around lab kill-switch controls… steady
    model-assessed
    ◆ 0 steady
  5. AI-enabled hackingAI-enabled hacking risk remains elevated as of Nov 14, unchanged since Nov 12 steady
    model-assessed
    ◆ 0 steady
The day, in more detail (5)
The day
  • Researched filings show Nvidia CEO Jensen Huang resigned Nov 11 amid an alleged Chinese intelligence targeting probe…
  • A researched forensic analysis published Nov 12 claimed six Chinese frontier models show Blackwell fingerprints…
  • Congress opened new Nvidia/H200 inquiries and locked in testimony from Huang…
  • The open-weights race and neocloud financing remained in stasis Nov 11-14 with no new frontier Apache-2.0 weight drop and no new Texas moratorium, project halt…

State of play · not in today's sections

Compute, chips and supply: export controls and China's compute gap · situation → , fast pace

Nvidia CEO Jensen Huang resigns as forensic claim ties banned Blackwell chips to Chinese models

The rundown

Since October 2022 the United States has tightened export controls on advanced computing and chipmaking tools to limit China's access to the most powerful AI chips, while China keeps buying tools and tries to build domestic alternatives. The supply chain runs from Nvidia's designs through TSMC's factories and packaging in Taiwan, where Korean stacked memory is integrated, before shipment to US buyers, so controls and smuggling shape who can train frontier AI models.

Stasis since Nov 9 broke. Nvidia CEO Jensen Huang resigned Nov 11 (8-K filed Nov 12) amid the alleged Chinese intelligence targeting probe, and a Nov 12 forensic analysis claimed export-banned Blackwell chips initialized six Chinese frontier models — a specific technical claim that does not by itself prove smuggling. Congress locked in testimony before Thanksgiving and opened new inquiries, while Malaysia opened a re-export probe. Confidence is high on the resignation, testimony commitments and forensic publication; the smuggling vs offshore-rental question remains unverified.

On the record

  • Jensen Huang · Chief Executive, Nvidia · 10 Nov

    “I support the investigation and will cooperate fully,”

    Commitment · conciliatory toward United States, Nvidia as reported by Deutsche Welle ↗

  • Jensen Huang · Chief Executive, Nvidia · 10 Nov

    “I am committed to protecting NVIDIA and supporting America's national security.”

    Reassurance · conciliatory toward United States, Nvidia as reported by Deutsche Welle ↗

  • United StatesDonald Trump · President of the United States · 10 Nov

    “I just know Jensen very well. I just think he’s a very honourable man,”

    Reassurance · conciliatory toward Jensen Huang, Nvidia as reported by South China Morning Post ↗

All on the record (5) · Compute, chips and supply: export controls and China's compute gap, across the day's updates
  • Jensen Huang · Chief Executive, Nvidia · 10 Nov

    “I support the investigation and will cooperate fully,”

    Commitment · conciliatory toward United States, Nvidia as reported by Deutsche Welle ↗

  • Jensen Huang · Chief Executive, Nvidia · 10 Nov

    “I am committed to protecting NVIDIA and supporting America's national security.”

    Reassurance · conciliatory toward United States, Nvidia as reported by Deutsche Welle ↗

  • United StatesDonald Trump · President of the United States · 10 Nov

    “I just know Jensen very well. I just think he’s a very honourable man,”

    Reassurance · conciliatory toward Jensen Huang, Nvidia as reported by South China Morning Post ↗

  • Nvidia · company, Nvidia · 9 Nov

    “target of a foreign intelligence effort”

    Policy · about Nvidia, China as reported by Deutsche Welle ↗

  • Nvidia board · Board of Directors, Nvidia · 8 Nov

    paraphrase The board currently has no reason to believe that US national security or Nvidia's confidential information had been compromised, but will examine all available information in its independent investigation

    Reassurance · about United States, China as reported by Anadolu Agency ↗

A statement shows what was said, not that it is true.

A statement shows what was said, not that it is true.

New · researched reported

  • Escalated: From steady flat stasis Nov 9-12 (pending board probe, pending House briefing, no BIS/seizure action) to fast escalation…
  • 13 Novtheguardian.com ↗
  • 12 Novthe-decoder.com ↗
  • techcrunch.com ↗
  • 11 Novsec.gov ↗

Watch for

  • Huang and Lutnick testimony before Thanksgiving (Nov 26) — scope of questioning on Megaspeed licensing and H200 tariff/revenue-share
  • Nov 18 joint House closed-door briefing on Huang interactions and Nvidia security protocols — whether it proceeds with interim leadership
  • Nov 17 Nvidia Q3 earnings — H200 China revenue, guidance, and any board probe update
  • Malaysia MITI findings on Megaspeed subsidiary re-exports and any BIS Entity List or licensing rule changes

Outlook Bipartisan pressure for tighter controls is likely to grow before Huang and Lutnick testify before Thanksgiving, with Chip Security Act (mandatory location verification) and Remote Access Security Act (cloud-rental loophole) as near-term legislative vehicles. Whether the Blackwell fingerprints reflect smuggling or lawful offshore rental is unlikely to be resolved before testimony, but either finding would intensify calls to close diffusion-rule and Entity List gaps. Market and governance volatility at Nvidia is likely to persist until a permanent CEO path and board probe findings are clarified.

Economies involved · size, trade, energy, arms · annual estimates

Open-weights race: Muse Spark Open, Nemotron, Beam and on-device models · situation → , flat pace

Open-weights stasis holds with no new frontier release as community ships incremental tooling

The rundown

The open-weights race pits openly released models — including Muse Spark Open, Nemotron and Beam — against closed frontier labs that keep weights private and sell access. 0; trillion-parameter models chase frontier capability while small models aim to run directly on phones and laptops, and who controls weights determines who can build.

Nothing has changed since Nov 11: open-weights remain in gradual catch-up to the closed frontier with high confidence. The last three days brought only incremental community work — llama.cpp v1.0 on Nov 13 — and no new frontier Apache-2.0 weight drop to close the gap.

On the record

  • Jensen Huang · Chief Executive, Nvidia · 10 Nov

    “I support the investigation and will cooperate fully,”

    Commitment · conciliatory toward United States, Nvidia as reported by Deutsche Welle ↗

  • Jensen Huang · Chief Executive, Nvidia · 10 Nov

    “I am committed to protecting NVIDIA and supporting America's national security.”

    Reassurance · conciliatory toward United States, Nvidia as reported by Deutsche Welle ↗

  • United StatesDonald Trump · President of the United States · 10 Nov

    “I just know Jensen very well. I just think he’s a very honourable man,”

    Reassurance · conciliatory toward Jensen Huang, Nvidia as reported by South China Morning Post ↗

New · researched reported

  • 14 Nov
  • 13 Novhuggingface.co ↗

Nothing changed since yesterday: Stasis held since Nov 13: no new frontier Apache-2.0 weight drop through Nov 14; only incremental community quantization work (llama.cpp v1.0 on Nov 13) continued, consistent with gradual catch-up pattern

Watch for

  • Independent reproduction of Ripple-65B 80.52 TerminalBench and Inkling 2 Small FrontierCode/DeepSWE scores and Hugging Face download/inference uptake
  • Inkling 2 Small Apache-2.0 weight release next week and external safety red-teaming results
  • Nvidia interim leadership statements on Nemotron roadmap and DOJ review of $12.9B Hugging Face acquisition

Outlook Open-weights will almost certainly remain in gradual catch-up near term absent a new frontier Apache-2.0 weight drop. Incremental quantization and on-device tooling are likely to continue but unlikely to shift the frontier gap on their own.

Economies involved · size, trade, energy, arms · annual estimates

Neocloud financing stress and data-center backlash · situation → , flat pace

Debt-funded neocloud build-out remains pressured with no new halt or financing relief confirmed

The rundown

The artificial intelligence buildout needs huge data centers to provide computing power. Specialist builders called neoclouds that rent out graphics chips and the labs that use them rely on Wall Street debt and on the power grid, which creates two pressure points: money gets tight when chip use wavers, and local communities can delay or block projects that raise electricity costs and land-use fights.

No new Texas moratorium, project halt or financing deal was confirmed since 2026-11-11, leaving the debt-funded neocloud build-out under the same community opposition, power-constraint and high-yield pressures seen since the Nov 6 selloff. Two reported items on Nebius spending plans and SoftBank gas losses did not change that standoff and no instrument moved outside its recent range. We are confident on stasis because no researched confirmation arrived.

New

    Nothing changed since yesterday: No new Texas statewide moratorium, Stargate Frontier (1,820 MW) or Nexus Hubbard (6,400 MW) halt/delay, or neocloud financing deal confirmed since 2026-11-11; pattern of community opposition, power constraints and high-cost financing held steady. Two reported headlines on Nebius $40B plans and SoftBank gas loss did not provide researched confirmation and no instrument moved outside range.

    Watch for

    • Texas legislature text or vote on statewide data-center moratorium and ERCOT audit outcome
    • Any halt or delay notice for Stargate Frontier (1,820 MW) and Nexus Hubbard (6,400 MW)
    • CoreWeave bond yields / 5-year CDS and hyperscaler Q4 capex guidance in 10-Qs for funding tightening

    Outlook Pressures from community opposition, power constraints and high-yield financing are likely to persist in the near term without a confirmed moratorium, halt or funding deal to break the standoff. A shift toward off-grid gas and prepayment funding is likely to continue as workarounds, but unlikely to ease local backlash quickly. Almost certain that financing stress remains elevated until a researched halt, moratorium text, or priced deal appears.

    AI safety governance, evaluation and catastrophic-risk regulation · situation → , gradual pace

    AI safety governance held steady with no new law, pledge or enforcement as oversight pressure continued

    The rundown

    Frontier AI governance pits the handful of companies building the most capable models — OpenAI, Anthropic, Google, Meta, xAI and Nvidia — against state and federal efforts to prevent catastrophic harms such as help with biological or chemical weapons, autonomous cyberattacks, or loss of control over AI systems. New York's RAISE Act and proposals like the federal FRONTIER Act try to set safety, transparency and oversight rules, while labs have made voluntary commitments to internal controls and independent evaluation.

    Pattern since Nov 10 held through Nov 14: steady, gradual escalation around lab kill-switch controls, DOJ and EU oversight pressure, and bioweapon warnings. No new law, Paris pledge update, Responsible Scaling Policy decision, or enforcement action was evidenced in the last three days; the one new item was an Epoch AI incident report illustrating agentic misuse risk, which does not change the governance picture.

    New · researched reported

    • New: Epoch AI Nov 13 incident report on agentic exploit attempts and OpenAI Nov 12 analysis on hidden chain-of-thought added evidence…
    • 13 Novepoch.ai ↗
    • 12 Novthe-decoder.com ↗

    Nothing changed since yesterday: No new law, Paris pledge update, Responsible Scaling Policy decision, or enforcement action evidenced since Nov 12; steady gradual escalation around kill-switch controls, DOJ/EU oversight pressure and bioweapon warnings held.

    Watch for

    • Anthropic update to Responsible Scaling Policy or formal RSI/2x decision, and any less-redacted METR evidence or independent replication of 1.9x estimate
    • Independent replication of Control Room coverage (unmonitored traffic share, RL monitoring) and of GPT-6 Astra covert-underperformance fix
    • EU or company disclosure responding to Commission Nov 9 request: incident counts, EU impact, and changes to risk assessments/security practices

    Outlook Steady, gradual escalation around kill-switch/monitorability and oversight pressure is likely to continue in the near term, with no new law or binding pledge expected imminently given Paris Call has no deadlines and RAISE enforcement portal only just opened. A shift would most likely come from an Anthropic Responsible Scaling Policy update or EU disclosure responding to the Nov 9 Commission request, but neither is evidenced as imminent.

    Economies involved · size, trade, energy, arms · annual estimates

    AI-enabled hacking and agentic cyber offense · situation → , flat pace

    AI hacking risk stays elevated as exposed agents persist and defenders scale automated patching without mass exploitation

    The rundown

    AI-enabled hacking pits frontier labs with the most capable closed models against open-weight and Chinese developers whose models are catching up on cyber tasks. Attackers seek cheap automated agents that can find and exploit flaws without human expertise, while labs and defenders try to contain that spread by building defensive factories that automate detection and patching. It matters because agentic cyber offense lowers cost and skill barriers to chaining reconnaissance, exploitation and persistence.

    AI-enabled hacking risk remains elevated as of Nov 14, unchanged since Nov 12. Exposed agentic systems remain publicly reachable and defender hardening continued with new automated patching products, with no new mass exploitation at scale confirmed beyond previously linked incidents.

    New · researched reported

    • New: Defender hardening advanced with OpenAI Daybreak Red Team/Defense Factory launch Nov 13…
    • 13 Novthe-decoder.com ↗
    • 12 Novtechcrunch.com ↗

    Nothing changed since yesterday: No new mass patch-gap exploitation at scale confirmed since Nov 12; pattern of elevated risk with exposed agentic systems and defender hardening held.

    Watch for

    • Censys or other scans showing Artex/Grok exposure count falling or evidence of exploitation of exposed consoles
    • Mozilla and Linux distro promotion of GPT-6 Cyber-found fixes to stable and time to reach most users
    • Evidence of GPT-6 Cyber or Volta-style LLM ranking used in wild exploitation or monetization beyond lab samples

    Outlook Risk likely to stay elevated in the near term. Defender factories will likely expand automated remediation, but the large pool of exposed agentic apps and AI infrastructure breaches keeps opportunistic exploitation plausible. Mass exploitation at scale remains unlikely without evidence of active weaponization of the exposed consoles, but roughly even that isolated abuse continues undetected.

    Also watching

    Quiet