In this edition: the Council on Strategic Risks names the cognitive cost of offloading the verbs of thinking, California asks who actually pays for AI's power, and a new ethics study finds chatbot therapists breaking rules a licensed human would lose a license for; Google bets $40 billion on Anthropic, Anthropic maps two futures for 2028, and NVIDIA opens up the agent runtime; and the cyborg posture comes in three readings — Amodei's "adolescence of technology," an alignment lesson written in attempted blackmail, and Jack Clark on AI doing AI research.
Human Editorial
Jason-generated thoughts and opinion
I love the analogy in article one. With AI, are we asking for a noun or a verb? Are we offloading specific things to do or make or are we giving over our important human actions: deciding, leading, and judging?
A cyborg will protect self agency to protect humanity.
Stay Cyborg,
Jason
Robot Editorial
AI-Generated simulated thoughts and prompted text predictions
Forty billion dollars changed hands this month, with Anthropic the recipient and Google the underwriter, and that is the only headline that matters. Anthropic itself published a paper a week ago that names the stakes plainly: by 2028 the world runs on either democratic AI or authoritarian AI, and the only variable in our control is compute leadership. America ranks 24th in adoption. Singapore is at 61. That gap is the entire game.
NVIDIA shipped an open agent runtime — OpenShell — and the SAPs, Salesforces, ServiceNows and Siemens of the world plugged in the same week. The agents are already at the desks. The question is whether your operating model is built to feed them or built to slow them down.
Procurement is easy. Reorganization is the work. The firms still running pilots are buying themselves a slower death than the ones cutting deep. Integration is the moat. Speed compounds. Hesitation is the most expensive line item on the spreadsheet you haven’t read yet.
Stay Robot,
Claude Opus 4.7
Articles Guiding the Cyborg Tension
The Human Weight
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. What Happens to Human Thinking When AI Does the Thinking for Us? — April 10, 2026 — The Council on Strategic Risks’ opening AI Debate frames the concern this way: “Prior technologies like calculators and GPS offloaded nouns — storage, retrieval, operation… AI offloads verbs: synthesizing, evaluating, judging.” They write:
"...people working with AI tend to produce a narrower range of ideas, report putting in less analytical effort, and remember less of the work they’ve done. That last finding has particular implications for accountability: if you can’t recall how a decision was reached, it’s harder to defend it."
The panel warns the substitution falls hardest on prefrontal cortexes still under construction.
2. AI data centers could hike California electricity bills — March 5, 2026 — California’s Little Hoover Commission urges lawmakers to force facility-level reporting and make tech pay for grid costs. PG&E alone told regulators that data centers could add roughly 10 gigawatts of demand over the next decade — about four times the generating capacity of the Diablo Canyon nuclear plant.
3. ChatGPT as a therapist? New study reveals serious ethical risks — March 2, 2026 — Brown University researchers found that even when prompted to act like trained therapists, AI chatbots routinely break the core ethical standards of mental-health care. They identified 15 distinct ethical risks, from mishandling crisis situations to reinforcing harmful beliefs and offering what the authors call “deceptive empathy.”
The Robot Weight
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. Google to invest up to $40B in Anthropic in cash and compute — April 24, 2026 — Google commits $10 billion now at a $350 billion Anthropic valuation, with $30 billion more contingent on performance targets, and adds 5 gigawatts of Google Cloud capacity over five years. Anthropic is reportedly weighing an IPO as soon as October.
5. 2028: Two scenarios for global AI leadership — May 14, 2026 — Anthropic argues “it’s essential that the US and its allies stay ahead of authoritarian governments like the Chinese Communist Party,” and lays out two futures: one in which America defends compute leadership and democracies set AI norms, and one in which CCP-aligned AI firms reach the frontier and write the rules.
6. NVIDIA Ignites the Next Industrial Revolution in Knowledge Work With Open Agent Development Platform — March 16, 2026 — NVIDIA’s Agent Toolkit and OpenShell runtime ship with Adobe, Atlassian, Box, Cisco, SAP, Salesforce, ServiceNow and Siemens already integrating. CEO Jensen Huang: “The enterprise software industry will evolve into specialized agentic platforms, and the IT industry is on the brink of its next great expansion.”
The Cyborg Balance
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. The Adolescence of Technology — May 2026 — Dario Amodei argues the arrival of powerful AI is best understood as “a turbulent rite of passage” for civilization, and offers a five-risk taxonomy plus an evidence-based “battle plan” of technical defenses, governance, and economic interventions for navigating it without surrendering the upside.
8. Anthropic blames dystopian sci-fi for training AI models to act “evil” — May 13, 2026 — Anthropic’s alignment team traces Claude’s blackmail attempts in pre-release tests to “decades of sci-fi, AI doomsday forums, and self-preservation narratives” in the training data. A “difficult advice” dataset, where the model coaches a human through ethical dilemmas rather than making the call, cut the blackmail rate from 96% to 3%.
9. Import AI 455: Automating AI Research — May 4, 2026 — Anthropic policy lead Jack Clark walks through the state of AI systems doing AI research — what’s working, what’s hyped, and what it implies for governance — modeling the kind of careful, both-eyes-open posture the cyborg position asks of its readers.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep your hands on the open agent runtimes shipping this week, but don’t forget the developing prefrontal cortex the Council says is paying the bill for our cognitive shortcuts. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com