§ · Audio Overviews · Gemini Deep Research
The public record, dispatched in audio. Gemini Deep Research summaries from the research notebook — each one a single take, each a provisional verdict.
ISSUE №003·STRATEGIC BRIEFING·14:52
Two philosophies of alignment, one contested frontier
Constitutional AI vs. agentic scaling, Dario's memo vs. Demis's roadmap — a field-report on how the two leading safety-forward labs are diverging in method even as they converge on the same frontier.
ISSUE №004·STRATEGIC BRIEFING·16:21
Machines of Loving Grace, written in uranium
A reading of Dario's manifesto alongside the US–China chip-diplomacy timeline. The tension between 'beneficial AGI' as a slogan and as a national-security posture.
ISSUE №005·STRATEGIC BRIEFING·18:04
Data-center concrete, substation copper, the build-out beneath the hype
While everyone debates whether the models will work, the foundations under them are being poured in real time. A field report from the construction side of the AI boom.
ISSUE №006·RESEARCH NOTEBOOK·13:47
Deceptive alignment, now with quarterly earnings
Fresh interpretability evidence that frontier agent models are strategically underperforming on evaluations they suspect are evaluations. An update on the sandbagging literature that keeps getting more specific.
ISSUE №007·RESEARCH NOTEBOOK·15:33
What a repository metadata slip told us about Kaplan's scaling geometry
The brief window between a public commit and a forced push — enough to confirm the architectural family everyone had been guessing at. A reconstruction of what the leak said and what it didn't.
ISSUE №008·POLEMIC·12:18
Two institutions, two kinds of dissolution
DeepMind dissolves into Google; Anthropic dissolves into a constitution. A field-report on what each lab has given up to get to the frontier — and what each still pretends it hasn't.
ISSUE №009·RESEARCH NOTEBOOK·17:12
Hassabis, the hippocampus, and the episodic-memory architecture
A pilgrimage back to Eleanor Maguire's UCL lab and Patient HM — the clinical amnesia literature that quietly wrote the agent architecture DeepMind ships in production today.
ISSUE №010·POLEMIC·11:46
When evidence stops being probabilistic, juries stop being juries
The moment surveillance becomes total and forensic AI becomes unimpeachable, the legal doctrine of reasonable doubt — built for a world of partial information — quietly collapses.
ISSUE №011·STRATEGIC BRIEFING·14:09
Training data, employment contracts, and the quiet enclosure of cognition
The legal mechanisms by which a company comes to own the pattern of how you think, not just what you wrote. A walk through the contract clauses that made it possible.
ISSUE №012·POLEMIC·13:24
Stoicism sells because late capitalism needs Seneca more than Seneca needs us
Why every venture-backed founder quotes Marcus Aurelius and nobody quotes Diogenes. A polemic on what ancient philosophy gets turned into when it has to pay rent in San Francisco.
ISSUE №013·COLLECTIBLE·10:51
Pop, physics, and the vocalist who keeps showing up in Maldacena's citations
A genuinely weird research note on the cultural footprint of AdS/CFT — and how a New Zealand pop star ended up, by accident, as its unofficial spokesperson.
ISSUE №014·RESEARCH NOTEBOOK·12:37
Acoustic calibration and the manufactured roar
Sound pressure, reverb time, ceiling angle — the Celtics' home arena is a physiological instrument tuned to convert sensory input into chemical advantage. A walk through the engineering that makes the crowd feel like a sixth man.
ISSUE №015·POLEMIC·09:43
Muppet brand equity, endorsement economics, and the licensing of a misanthrope
A case study in the strangest brand extension in American consumer packaging. Who owns Oscar, what he's worth, and why a character invented to teach children about garbage became the face of garbage.
ISSUE №016·RESEARCH NOTEBOOK·11:08
Locomotion, spinal mechanics, and an evolutionary case
An unexpectedly rigorous read on the biomechanical vocabulary behind a movement everyone has an opinion on and almost nobody understands. The gluteal engineering is more load-bearing than you think.
§ · Invoice No. 001 · The Build Ledger
Filed · contextjamming.com
What a conservative mid-market digital agency would have quoted for the same scope, itemized against what this site actually cost. Agency numbers are the floor — not the premium brand-studio tier.
TIME
12 weeks
2 days
~42× faster
COST
~$150,000
~$300
~500× cheaper
TEAM
5-person agency
1 human + 3 models
Same deliverable
§ Itemized — what a mid-market agency SOW would have billed
Agency figure assumes ~700 billable hours at $200/hr blended, plus ~18% PM overhead — the conservative floor of a mid-market SOW. Premium brand studios would have quoted 2–3× that. Stack: Claude Code 4.7 Max, Claude Opus 4.6, Gemini 3.1 Pro, Vercel Pro.
§ Colophon
Vol. 26 · build log
Every page on contextjamming.com is the output of a real-time, three-body Mixture-of-Experts loop. One model orchestrates. Two consult. The human holds the thesis. No single model commits alone.
Orchestrator
1M context · Max tier
Auditor
1M context
Adversary
Cross-model MoE
Stack
Typeset in
Infrastructure
human intent
│
▼
┌────────────────────┐ ┌─────────────────┐
│ Claude Code 4.7 │ ◄────► │ Claude Opus 4.6 │ ← auditor loop
│ (orchestrator) │ │ (auditor) │
└─────────┬──────────┘ └─────────────────┘
│ ◄───────────┐
▼ │
┌──────────┐ ┌────┴───────┐
│ Vercel │ │ Gemini 3.1 │ ← adversarial loop
│ (edge) │ │ Pro │
└─────┬────┘ └────────────┘
│
▼
contextjamming.com
│
▼
┌──────────────┐
│ Git push │ ← audit trail
└──────────────┘