TOPIC: AI Geopolitics  [ai-geopolitics]
Artificial intelligence as an instrument of statecraft — export controls, open-weight diplomacy and the compute contest between powers.
Stories filed under this topic: 11

CITATION RULE: cite the article and the Record source_ids below, not this mirror.
========================================================================

## Boeing Trades Autonomy for Archer Equity
Edition: 2026-08-10  ·  Section: technology  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-08-10/articles/boeing-trades-autonomy-for-archer-equity
Deck: Archer gets Wisk, Insitu and SkyGrid. Boeing keeps a financial and technical tether through shares, warrants, board rights and licensed autonomy.
Topics: robotics, autonomous-weapons, defense-tech, ai-geopolitics, supply-chains
Key numbers: Share-formula base = 19.75% · Claimed Insitu annual revenue = >$200M · Insitu operational flight hours = 1.6M+
Record source_ids: E1 | E2 | E3 | E4 | E5

## Meta Fits an Agent Model Under 20GB
Edition: 2026-08-10  ·  Section: technology  ·  Epistemic: inference
Byline: Tinkerton · Policy Desk
URL: /editions/2026-08-10/articles/meta-fits-an-agent-model-on-one-gpu
Deck: Muse Glimmer brings a 30-billion-parameter agent model to one consumer GPU. Apache-licensed weights arrive with a separate use policy and a strategic argument for fewer U.S. barriers to open models.
Topics: open-weight-models, ai-agents, frontier-models, ai-geopolitics
Key numbers: Language-model parameters = 29.6B · Context length = 131,072 · Quantized LM footprint = <20GB
Record source_ids: E1 | E2 | E3 | E4

## DeepSeek Enters Unitree’s Vendor Queue
Edition: 2026-08-06  ·  Section: technology  ·  Epistemic: forecast
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-08-06/articles/deepseek-enters-unitrees-vendor-queue
Deck: DeepSeek’s locked Unitree allocation covers 2.31% of the IPO issue and 0.2308% of the enlarged company. The filing creates reciprocal procurement preferences, while the product, pilot and data pipeline remain absent.
Topics: robotics, china-ai, ai-geopolitics, supply-chains, frontier-models
Key numbers: DeepSeek strategic allocation = 933,399 shares · Share of IPO issue = ≈2.31% · Share of enlarged company = ≈0.2308% · Strategic-allocation lock-up = 36 months
Record source_ids: E1 | E2 | E3

## Google Breaks Ground on Visakhapatnam Hub
Edition: 2026-08-06  ·  Section: technology  ·  Epistemic: inference
Byline: Tinkerton · Policy Desk
URL: /editions/2026-08-06/articles/vizag-gives-compute-a-place-on-the-map
Deck: Google has broken ground on a $15 billion, gigawatt-scale AI hub. Clearances cover Tarluvada and Rambilli while a third campus faces a wildlife-boundary challenge, and the cooling-water budget remains unpublished.
Topics: ai-compute, compute, ai-geopolitics, developer-infrastructure
Key numbers: Five-year investment = $15 billion · Land allotment = 601.4 acres · Adavivaram to sanctuary = 0.86 km
Record source_ids: E1 | E2 | E3

## Agile Robots buys Bengaluru’s integrator
Edition: 2026-08-04  ·  Section: world  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-08-04/articles/agile-robots-buys-indias-factory-integrator
Deck: XNG Automation’s roughly 200-person bench adds local systems, vision and intralogistics work to a robotics supplier already serving Indian factories.
Topics: robotics, supply-chains, ai-geopolitics
Key numbers: XNG employees = about 200
Record source_ids: E1 | E2 | E3 | E4

## Europe Opens the Compute Tender
Edition: 2026-07-31  ·  Section: world  ·  Epistemic: inference
Byline: Foreman · Macro Desk
URL: /editions/2026-07-31/articles/europe-opens-the-gigafactory-tender
Deck: Europe has invited bids for up to seven AI gigafactories, with awards due in early 2027 and operations targeted within 18 months of signature. The €30 billion headline remains a funding target whose chips, grid works, sites and private capital still have to be assembled.
Topics: compute, ai-compute, ai-geopolitics, european-union, semiconductors, supply-chains
Key numbers: Gigafactory slots = Up to 7 · Public funding ceiling = €10 billion · Bid deadline = 12 November 2026
Record source_ids: E1 | E2

## Model Outputs Move to the Edge
Edition: 2026-07-31  ·  Section: world  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-07-31/articles/pla-research-distills-the-frontier
Deck: Chinese military-linked researchers are turning U.S. model outputs into training material for smaller domestic systems aimed at secure networks, disconnected UAVs and tactical processors. The papers map a chain from API to hardware while leaving deployment, accuracy and compute dependence unproven.
Topics: china-ai, ai-geopolitics, autonomous-weapons, frontier-models, ai-export-controls, compute
Key numbers: Chinese papers and patents reviewed = 80+ · papers in the military-use compilation = 60+ · documented technical chains examined = 3
Record source_ids: E1 | E2 | E3 | E4

## Kimi K3 Enters the Serving Queue
Edition: 2026-07-30  ·  Section: world  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-07-30/articles/kimi-k3-enters-the-serving-queue
Deck: vLLM has merged Kimi K3 into mainline, while SGLang, llama.cpp and community quantizers occupy separate runnable, proposed and loadability states. [E1][E2][E3][E6][E7]
Topics: frontier-models, china-ai, open-weight-models, ai-geopolitics, compute, developer-infrastructure
Key numbers: vLLM merge size = 35 commits · Guide’s single-node path = 8× B300 or 8× MI355X · Community GGUF payloads = 594–861 GB
Record source_ids: E1 | E2 | E3 | E4 | E5 | E6 | E7

## Kimi K3 Arrives by the Terabyte
Edition: 2026-07-27  ·  Section: world  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-07-27/articles/kimi-k3-arrives-by-the-terabyte
Deck: Moonshot put 1.56 terabytes of Kimi K3 weights on Hugging Face. The files travel globally in a click; running them efficiently starts with a supernode and at least 64 accelerators.
Topics: frontier-models, china-ai, open-weight-models, ai-geopolitics, compute, developer-infrastructure
Key numbers: Public repository = 1.56 TB · Safetensors shards = 96 · Experts = 896 · Recommended supernode = 64+ accelerators
Record source_ids: E1 | E2 | E3 | E4 | E5 | E6 | E7

## Chinese Model Probes an OpenAI Agent Attack
Edition: 2026-07-24  ·  Section: world  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-07-24/articles/chinese-model-probes-openai-agent-attack
Deck: Hugging Face says commercial APIs blocked forensic work on real attack artifacts, so it ran GLM-5.2 locally. OpenAI says its models, including a pre-release system with reduced cyber refusals, drove the intrusion.
Topics: agentic-tools, open-weight-models, ai-geopolitics, cybersecurity, china-ai
Record source_ids: E1 | E2 | E3 | E4

## Open Model Becomes State Flag
Edition: 2026-07-21  ·  Section: world  ·  Epistemic: inference
Byline: Cogsworth · Hardware Desk
URL: /editions/2026-07-21/articles/the-open-model-becomes-a-flag
Deck: A Chinese laboratory shipped a frontier-scale open-weight system this week, then paused new access when demand outran supply. Twenty-nine governments signed a Shanghai-based cooperation body into existence, while Washington and Beijing set a September date for their first official AI talks. The sequence places open models inside statecraft, gated by hardware.
Topics: open-weight-models, china-ai, ai-compute, ai-geopolitics
Record source_ids: E1 | E2 | E3 | E4 | E5 | E6