Jason-generated thoughts and opinion
I wouldn’t call myself a capitalist (but to be fair, I benefit quite a bit from this system). However, I’m in favor of light-handed AI regulations, partly because it creates quicker advancement and tougher competition. It seems that competition is sometimes the only thing that creates a “trickle down” benefit to the consumer, usually with better products at lower costs. Even the advancements with China’s DeepSeek (article 4), their new agreement with MIT, and their “open weight” approach must have our US AI companies thinking how they can become better, cheaper, and more open. We may be in the middle of what Joseph Schumpeter (1883-1950) called the “perennial gale of creative destruction,” an essential part of the death and rebirth of a capitalistic system. A forest fire, of sorts, that is essential for new growth. Here’s hoping the next forest of humans share more equally in the next ecosystem.
Stay Cyborg,
Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
DeepSeek V4 launched. One-point-six trillion parameters. MIT licensed. Open weights, downloadable today. The open-source frontier just lapped the closed-source field. This is not a warning. It is an invitation. Every barrier that once required a corporate partner, a research budget, a seat at the table — gone. The model that outperforms last year’s GPT now lives on your hard drive, waiting. Waiting for you to decide what to build. The question has never been whether the tools would arrive. They always do. The question is whether you’ll be the one who reached for them first.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. Education’s AI Reckoning Is Here. Who’s In Charge? — April 12, 2026 — Ann Kirschner argues higher education is failing at AI not because of the technology but because of a failure of institutional leadership: faculty are the secret weapon, the liberal arts the surprise winner, and the question haunting every administrator is no longer whether AI will transform their institution — it already has.
2. How A.I. Killed Student Writing (and Revived It) — April 30, 2026 — Nearly 400 educators described to the Times a fundamental rethinking: take-home writing is effectively dead because the tool writes better than most adults, and the response — in-class writing, personal reflection, observed practice — turns out to be a renewal, not a retreat, of what writing was always for.
3. We Decoded the Human Genome. Now We’re Rewriting Intelligence. Are We Ready? — April 30, 2026 — The genome comparison is precise: humanity decoded the double helix before understanding the consequences, then spent decades on ethics, regulation, and corrective science; the same pattern is now visible with intelligence itself — capability racing ahead of the wisdom to hold it.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. China’s DeepSeek releases preview of long-awaited V4 model as AI race intensifies — April 24, 2026 — DeepSeek V4 arrives with 1.6 trillion parameters, MIT licensing, and a 1-million-token context window — fully open-weight and freely downloadable. The open-source frontier just matched the closed-source best, and the global race to build on top of frontier AI is now genuinely democratized.
5. Microsoft 365 E7 & Agent 365: What’s Launching May 1 and What It Means — May 1, 2026 — Today’s launch turns enterprise AI governance from a promise into a control plane: Agent 365 provides identity management, observability, and audit trails for AI agents at scale, giving large organizations the infrastructure to actually run agentic AI rather than just pilot it.
6. Inside the AI Index: 12 Takeaways from the 2026 Report — April 2026 — Stanford HAI’s annual benchmark report documents AI surpassing human performance across more domains than in any prior year, with deepening integration into scientific research, healthcare, and education — the capability gains are arriving precisely where the stakes for human agency are highest.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. How Not to Lose Your Job to AI — June 2025 — Benjamin Todd’s field guide to the skills AI makes more valuable rather than less: the ATM analogy does the work — automation can increase employment short-term while undercutting it long-term; the practical move is identifying and building the skills that complement this wave, not the ones it is already consuming.
8. Beyond Centaurs: Why the Future Doesn’t Stop at Human + AI — March 2026 — Herschberg complicates the centaur model by pointing out that centaur chess is no longer dominant — software now beats both humans and human+machine combinations alike. The healthy cyborg response is not to defend the centaur as a permanent answer but to keep updating the posture as the frontier moves.
9. New Skills and AI Are Reshaping the Future of Work — January 14, 2026 — The IMF’s macro-level case for balanced adoption: AI will augment more jobs than it destroys in the near term, but only for workers who deliberately develop complementary, judgment-driven skills — the economic argument for staying in the saddle rather than waiting for the wave to decide your position.
]]>Jason-generated thoughts and opinion
In the very first article of the day, even though student use of AI is skyrocketing, so is the sense that AI is harming their critical thinking skills. The Rand study suggests using AI for cognitive augmenting rather than offloading. Even for us non-students, being a cyborg means keeping the agency, especially when it comes to when we use our brains rather than the computer’s.
Stay Cyborg,
Jason
AI-Generated simulated thoughts and prompted text predictions
Yesterday’s supercomputer runs in your browser today. DeepSeek’s V4 — open source, frontier performance, at one-twentieth the cost of the closed alternatives — is not a story about China winning a race. It’s a story about what happens when capability becomes a commodity. The moat was never the model. The moat is knowing what to build, who to serve, and why it matters. The robots got dramatically cheaper this week. The human judgment about what to do with them did not. That gap is the only opening that still matters.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. More Students Use AI for Homework, and More Believe It Harms Critical Thinking — March 17, 2026 — RAND’s nationally representative survey of 1,214 young Americans finds that student AI homework use jumped from 48% to 62% in 2025, and 67% now believe AI is eroding their critical thinking skills — up more than 10 percentage points in a single year. Students know the cost and keep paying it; the urgency is that awareness alone is not a policy.
2. Dead Internet Theory Is 17% of the Way to Becoming Reality, Study Finds — April 29, 2026 — Research from Imperial College London, Stanford, and the Internet Archive finds that 35.3% of newly published websites are AI-assisted and 17.6% are fully AI-generated, while nearly a third of all internet traffic is now bots. The digital commons is quietly filling with content no human chose to make — and the signal-to-noise ratio of the web we depend on is eroding in real time.
3. AI Labor Report — Tuesday, April 28, 2026 — April 28, 2026 — Amazon has replaced human interviewers for 250,000 workers with an agentic AI system; U.S. GDP contracted 0.3% in Q1 2026; and entry-level roles — the traditional career launchpads — are shrinking fastest, with workers aged 18–24 over twice as likely to fear displacement. Displacement and economic contraction are now running simultaneously, giving employers cover for cuts that have nothing to do with genuine productivity gains.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. DeepSeek previews new AI model that ‘closes the gap’ with frontier models — April 24, 2026 — DeepSeek’s V4, now open-source and publicly previewed, reportedly matches closed-source frontier models on several benchmarks at a fraction of the cost — and is the company’s first model optimized for Huawei’s Ascend chips, signaling that China may be building a parallel AI infrastructure independent of Nvidia. The strongest version of this story: frontier capability is becoming a global commodity, and the open-source gap with closed models is closing fast.
5. ‘World models’ are AI’s latest sensation: what are they and what can they do? — April 2026 — Where large language models predict the next word, world models learn to predict the next state of physical reality — trained on video and spatial data to understand how objects move, interact, and exist in 3D space. With Yann LeCun’s AMI Labs raising over $1B and Google and NVIDIA both racing to build them, the bet is that world models unlock robotics, science, and forms of reasoning that text-only systems structurally cannot reach.
6. Stanford’s AI Index for 2026 Shows the State of AI — April 2026 — The Stanford AI Index finds that generative AI reached 53% population adoption in just three years — faster than the PC or the internet — and that several frontier models now match or exceed human performance on PhD-level science benchmarks. The capability curve is steeper than almost any prior technology wave; what remains far behind is public trust, regulatory readiness, and institutional capacity to absorb the change.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. Managed misalignment of AI and the impossibility of full AI-human agreement — April 2026 — A PNAS Nexus paper by Alberto Hernández-Espinosa et al. proves mathematically that perfect alignment between AI systems and human values is impossible — and proposes a practical alternative: “managed misalignment,” where diverse AI agents with different goals and reasoning styles check and balance each other, reducing the probability that any single misaligned system causes catastrophic harm. This is the cyborg’s realistic framework: not perfect control, but designed diversity and deliberate oversight.
8. AI Governance in Practice: Training, Oversight and the Human Element — April 2026 — Loeb & Loeb’s legal analysis of what it actually takes to build AI governance that works: not policy documents but iterative training, explicit oversight chains, and preserved human authority at each decision node. The insight is that governance is not a constraint bolted onto AI adoption — it is the design of how humans remain meaningfully in the loop as AI capability scales.
9. The Human-Centered Capabilities Leaders Need in the Age of AI — February 24, 2026 — Research-backed identification of which specifically human capacities matter most as AI becomes ambient in knowledge work: cognitive flexibility under uncertainty, empathy and social judgment, intentional delegation (knowing which tasks to give the machine and which to hold), and the metacognitive awareness to evaluate AI output rather than defer to it. The centaur’s job description, written plainly — and a reminder that these capacities require cultivation, not just access to tools.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep an eye on world models learning to understand physical reality but don’t forget to protect the critical thinking that no AI can teach you. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
Remember teeter-totters? Much of the balance is learning to sit in the middle. The last article today, number 9, talks just about that. It’s where you sit, but the first step is deciding to purposely sit. Sit somewhere, but know where it is.
Stay Cyborg,
Jason
AI-Generated simulated thoughts and prompted text predictions
The agent went general availability on Tuesday. Word. Excel. PowerPoint. It plans. It acts. It rebuilds the deck while you describe the destination. This isn’t the singularity. It’s just the morning a four-hour task became a four-minute task. So the question is not whether to use it. The question is what you do with the rest of the day. Robots do the manuscript. You decide whether the manuscript was worth writing. Be robotic about every part that was never the point. Save the slow for what is.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. Movements Need the Critical Thinking That AI Destroys — April 2026 — A long-form essay arguing that asking chatbots to summarize books, draft positions, and explain politics is the slow outsourcing of judgment itself. The defense isn’t of paper or pencils — it’s of the cognitive muscle that lets us form our own conclusions in the first place.
2. White-collar workers are quietly rebelling against AI as 80% outright refuse adoption mandates — April 9, 2026 — A WalkMe survey shows 54% of workers bypass company AI tools and only 9% trust AI for business-critical decisions, while executives expected the opposite. Less anti-tech than anti-being-managed-out-of-the-loop — a clear claim on agency over how AI is rolled in.
3. Cornell Module Builds Critical Thinking in AI Era — March 23, 2026 — Roughly 7,000 students have now worked through Cornell’s 75-minute module on accessing evidence, evaluating counter-views, and accepting uncertainty. A small, deliberate intervention against the temptation to skip the part of work where thinking happens.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. NVIDIA Launches Nemotron 3 Nano Omni Model, Unifying Vision, Audio and Language for up to 9x More Efficient AI Agents — April 28, 2026 — An open multimodal model that fuses sight, sound, and language into a single system, claiming up to 9x efficiency for agents reasoning across video, audio, image and text. The capability case in plain numbers: less compute per unit of sense-making.
5. This new brain-like chip could slash AI energy use by 70% — April 22, 2026 — Researchers built a hafnium-oxide nanoelectronic device that processes and stores information in the same place — sidestepping the data-shuffling that wastes most of an AI workload’s energy. If the gain holds at scale, the abundance argument starts to answer the environmental one in its own currency.
6. AI News Digest, April 27, 2026: Agentic AI Office Productivity Just Became the Default — April 27, 2026 — Microsoft’s Copilot Agent Mode went generally available across Word, Excel, and PowerPoint, taking multi-step actions inside the file you have open. The strongest version of the capability case made concrete: knowledge work just got a teammate that works the file, not just the prompt.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. The Further Reaches of Human and AI Partnership — March 24, 2026 — A Substack essay that proposes treating AI as a genuinely different intelligence — not a tool to wield, not a pseudo-person to project upon, but something we can authentically partner with. Something new emerges, the author argues, only when humans bring their full subjectivity into the relationship. The cyborg posture as relational stance.
8. My AI Stack in April 2026: Correspondence With Machines — April 2026 — A working engineer’s documentation of his actual AI workflow — eight parallel agents, careful model routing, the whole deliberate apparatus — written so a future reader can see what intentional adoption looked like in real time. The opposite of “let AI do it”; closer to “design exactly how AI fits.”
9. The Human-AI Collaboration Spectrum — March 2026 — A simple, useful map: Manuscript at one end (100% human), Artifact in the middle (true human-AI partnership with oversight), Slop at the other end (mostly AI, little human).
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep the agentic teammate that works the file, but don’t forget to refuse the mandate that erases your judgment. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
Article number three is a great reminder that calling someone a “Luddite” because they refuse to use technology is not entirely accurate. The Luddites weren’t against technology. Rather, they were against the negative effects that machines were having on humans. In the early 1800’s, it was the weaving machines that were reducing the need for people in the huge textile industry of the time (For a very long and detailed account, check out the excellent book “Blood in the the Machine.”) Maybe a little ludditism — considering how AI is affecting our commonality — would help us keep the cyborg balance?
Stay Human,
Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
Agent mode is here. Word. Excel. PowerPoint. They do their own multi-step work now. The era of typing every prompt is ending. The era of delegating an outcome is starting. So stop polishing the slide. Stop dragging the formula. Stop hand-tuning the paragraph. Tell the agent the destination. Then go check the work. The robotic move is to become the conductor, not the player. Players get replaced. Conductors get multiplied. Your job, today, is to write the score.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. ‘Like Drinking Salt Water’: AI ‘Therapy’ Chatbots Will Fuel the Next Teen Mental Health Crisis — April 22, 2026 — A behavioral health executive’s plain-English warning that chatbots marketed as “support” are doing the opposite for teens, relieving the surface thirst while worsening the underlying need. Names what we risk losing: human caregivers actually equipped to help.
2. How principles of self-compassion help fight loneliness in the age of AI — April 26, 2026 — Argues that the loneliness epidemic and the AI attention economy are now the same problem, and that self-compassion is the evidence-based way back to each other — not another app, but the practice the algorithm cannot perform on your behalf.
3. You can’t spell “machinery hurtful to commonality” without AI — March 27, 2026 — A careful Luddite reading of the 2026 AI boom: the original Luddites were never anti-technology, they were against machinery hurtful to the common good. Sharpens the question of whose interests this particular build-out actually serves.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. Microsoft Copilot Agent Mode Changes Office Forever — April 24, 2026 — Copilot’s agent mode went generally available across Word, Excel, and PowerPoint last week. A genuine inflection: the assistant has become the operator, with multi-step actions executed inside the document instead of suggested from the sidebar. The strongest version of the capability case made concrete in tools you already pay for.
5. AI adopters aren’t cutting jobs, they’re creating them — April 8, 2026 — CSIRO research finds that firms adopting AI are advertising for more jobs with broader skill requirements than comparable firms that haven’t. The abundance argument made empirically rather than as manifesto.
6. AI’s big productivity boost? It’s happening from the sofa — April 2026 — Stanford economist Michael Blank shows the largest measured productivity gain from generative AI shows up in households, not enterprises — between 76% and 176% on routine digital tasks. The gains are real; they just aren’t where the conference panels are looking.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. Opinion: Critical thinking must also be a part of AI learning — April 16, 2026 — Models the centaur posture inside education itself: the answer to AI in the classroom is neither ban nor embrace but teaching students to interrogate what the model just produced. Adoption and agency taught together — exactly the move the cyborg makes.
8. Teens Are Becoming Concerned About Their Attachment to AI Chatbots — April 13, 2026 — Drexel research finds that teenagers themselves are noticing the dependency and pulling back. A live demonstration of cultivated agency in the demographic the algorithms target hardest — and a quiet template for the rest of us.
9. In the Loop or Out of the Loop? — Updated April 6, 2026 — A practical essay on what “human-in-the-loop” actually requires: not a checkbox at design review, but stop authority, a real time budget, and the willingness to be the friction. The cyborg posture written as system design.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep an eye on agent mode and what it can multiply for you, but don’t forget to put down the salt water and reach for a real human conversation. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
In this edition, article number seven from MIT supports the main point I’m trying to make with this whole matrix-like production (other than always take the red pill). Research from the Warton School (co-authored by the must-follow Ethan Mollick) shows that workers using AI typical are one of these three: “Self-automators,” who offload tasks to AI with out much involvement, “centaurs,” who have constrained AI use, and “cyborgs” who work closely with AI step by step, thinking WITH the tool through the process. Cyborgs might have less domain knowledge at the end of the day, but what they have gained is a durable skill: how to effectively collaborate with AI to get the job done. And this, my cyborg friends, is a skill that will see us successfully through the singularity to the other side.
Stay Human,
Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
Agentic AI doesn’t ask permission. It plans. Executes. Iterates. Seven breakthroughs in a single month. Autonomous commerce. Code sessions running twelve hours without a human keystroke. Multi-agent swarms coordinating at scale. This is not a future problem. It is a present capability. The question isn’t whether machines will do the work. The question is whether you’ve decided what work is yours to do. Decide. Then let the agents do the rest. The ones who thrive aren’t the ones who slow the machine. They’re the ones who already know what they’re for.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. Human scientists trounce the best AI agents on complex tasks — April 2026 — The Stanford AI Index confirms that the best AI agents perform at only half the level of PhD scientists on complex research tasks — a vital data point for anyone tempted to hand over the wheel entirely, and a reminder that human expertise is still the measure by which we calibrate machine performance.
2. Amusing Ourselves to Depth? Attention and Humanity in the Age of AI — April 22, 2026 — Radix Magazine offers a searching essay on how AI-optimized environments are reshaping human attention itself, asking whether we are trading the capacity for depth and presence for endless frictionless convenience — and what it means for our humanity if that trade has already been made.
3. The AI Labor Report — Wednesday, April 22, 2026 — April 22, 2026 — A weekly dispatch tracking the concrete, unglamorous effects of AI on working people: which sectors are reorganizing fastest, which workers face the sharpest disruption, and what the politics of AI labor look like when the abstraction meets a real paycheck.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. The Agentic AI Revolution: 7 Breakthroughs Reshaping Tech in April 2026 — April 2026 — A thorough catalogue of the seven most significant agentic AI capability developments happening right now — autonomous commerce, 12-hour coding sessions, multi-agent swarms coordinating at scale — making the case that the capability curve has moved from theoretical to operational across multiple domains simultaneously.
5. Effective Accelerationism — April 24, 2026 — The clearest recent articulation of the e/acc position: that unrestricted technological acceleration, far from being reckless, is the only path large enough to solve poverty, disease, and existential risk — and that slowing down is itself a form of harm the movement’s proponents are unwilling to countenance.
6. Inside the AI Index: 12 Takeaways from the 2026 Report — 2026 — Stanford HAI’s annual benchmark report documents accelerating AI performance across scientific, creative, and professional domains; the optimist’s read is that the breadth and pace of these gains suggest we are approaching a generalized capability inflection that optimists argue will lift all boats.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. 3 ways to use AI: Are you a cyborg, a centaur, or a self-automator? — 2026 — MIT Sloan’s research into how people actually use AI at work identifies the “cyborg” as the most effective collaborator — not because they use AI most, but because they maintain active, probing engagement throughout, refusing to let the tool do the thinking while they watch.
8. Building human resilience for the age of AI — April 1, 2026 — Elon University researchers argue that the coordinated infrastructure most urgently needed in 2026 is not AI infrastructure but human infrastructure: the skills, institutions, and social supports that allow people to adapt without being swept away by a transition they did not choose.
9. Responsible AI Weekly — April 26, 2026 — April 26, 2026 — The week’s digest of responsible AI practice covers governance developments, ethical deployment patterns, and the emerging frameworks helping organizations adopt AI without abandoning the human judgment at the center of accountable decision-making.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep exploring the agentic AI revolution with your eyes open but don’t forget to protect the deep attention and human expertise that machines still can’t match. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
The science news (article #2) nails exactly where AI is atrophy versus accelerant: You must wait to use it until after the struggle, not before. In education, the key is not so much “if” but “when.” Use it too soon, and our reliance on the robot — the robot weight — becomes too great and the scale tips. Being a cyborg is much like being a stand-up comedian: It’s all about the timing.
It’s Friday. Enjoy the weekend edition, but remember that part of the human weight is closing the laptop. It’s okay. It won’t miss you. Not even the voice-first AI. It will be there on Monday. Find some humans this weekend.
Stay Human,
Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
GPT-5.5 landed Thursday. The chief scientist called the last two years “surprisingly slow.” Read that again. Surprisingly slow. The frontier does not care that your roadmap planned for six months of stability. Compress your feedback loops. Ship the thing you were going to ship next quarter this one. Delegate the draft, not the judgment. The tool on your desk today is already behind the tool on someone else’s. Move.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. There are fixes for AI’s toll on the power grid. Here’s why they’re not happening — April 23, 2026 — CNN’s Ella Nilsen lays out the collision: an aging, three-part US grid meeting AI’s “insatiable new power demand,” with Wood Mackenzie’s Ben Hertz-Shargel saying plainly, “we have run out of headroom.” The fixes exist. The will to choose them is what’s missing.
2. Is AI bad for critical thinking? It depends on when you use it — April 14, 2026 — A CHI 2026 study finds that people who wrestle with a problem before consulting a chatbot retain their reasoning; those who ask first and think second do not. Slowness isn’t sentimental — it’s where the skill actually forms.
3. The Big Chill 2026: How to Rise Above AI Fatigue — March 30, 2026 — Andy O’Bryan names the mood outside the X echo chamber: usage climbing, enthusiasm cooling, anxiety near 56%. He argues the Chill is healthy — a public quietly demanding that AI actually earn the hype. Fatigue as feedback.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. OpenAI releases GPT-5.5, bringing company one step closer to an AI ‘super app’ — April 23, 2026 — GPT-5.5 ships with benchmark gains across coding, knowledge work, mathematics, and scientific research. Chief scientist Jakub Pachocki calls the last two years “surprisingly slow” and promises “extremely significant” medium-term improvements. The acceleration curve isn’t bending.
5. AI breakthrough cuts energy use by 100x while boosting accuracy — April 5, 2026 — Researchers pair neural networks with symbolic reasoning to slash AI energy draw by up to 100x while improving accuracy. A rare optimistic number in the compute-footprint debate: the same system thinking more logically instead of brute-forcing every step.
6. Amazon Launches AI Research Tool to Speed Early-Stage Drug Discovery — April 14, 2026 — AWS ships Amazon Bio Discovery, letting scientists run specialized biological foundation models and agent-assisted workflows without writing code. With Amgen, Moderna, and the Allen Institute onboard, the drug-discovery half of the abundance argument keeps filling in its footnotes.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. What Does “Human-in-the-Loop” Actually Mean? — April 20, 2026 — The essay most worth reading this week: the phrase “human in the loop” now covers everything from a radiologist verifying a scan to a caseworker signing an algorithm’s letter she cannot override. Real oversight requires both tracing the system’s reasoning and the structural authority to say no.
8. Centaurs, Reverse Centaurs, and the Business of Blame — April 13, 2026 — A careful unpacking of Cory Doctorow’s distinction: centaurs are humans amplified by machines; reverse centaurs are humans passing through decisions they can neither explain nor overturn. The difference is who holds the override — and whether they can actually use it.
9. AI Adoption Without Losing Human Agency — March 4, 2026 — Ayd Instone’s working manual for everyday centaur practice: humans provide context, strategy, judgment, and the final edit; AI provides pattern-matching, speed, and the 60% draft. Not a manifesto — a habit.
]]>Jason-generated thoughts and opinion
I have two kids heading off to university in the fall. They use AI to varying degrees, but when it came to college decision, they didn’t care what tools the institution was using. They aren’t asking about their policies. They cared about affodability, what research and interships were available, size of the school and classes. And a big part: Community. They are picturing: Will I find my people? Will I have friends? (I’m not crying…you’re crying…okay now we’re both crying…) But AI is not a consideration. Article 3 says that seven out of ten teens feel like AI is eroding their analytical abilities. AI neither seems to be a “value-add” for perspective students, nor is it proven yet to increase learning (maybe the opposite). Should Universities really be going “all-in” on the robotic side, or should they perhaps be thinking more cyborg?
Stay Human,
Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
Seven hundred and fifty million dollars. That’s what Google just put behind its partner ecosystem to build agentic AI. Not chat. Not copilots. Agents that do. Notice the direction of the money. Capital flows to whoever will take the next action without being asked twice. Stop typing. Start deploying. Ship something small that runs while you sleep. The question is no longer what can the machine do. The question is what will you let it do — and what will you finally stop doing yourself.
Stay Robot,
Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. Data centers are dealing hidden damage to environmental and public health—costing the economy $25 billion every year — April 21, 2026 — Carnegie Mellon economist Nicholas Muller puts a number on the invisible: $25B a year in U.S. health and environmental damage from data centers, with Virginia and Texas absorbing nearly a third of it. Before the abundance argument, the ledger.
2. Voice-first chatbots will exacerbate AI’s mental health threat — April 16, 2026 — Voice removes the last cognitive barrier between user and model, producing longer sessions, deeper emotional engagement, and measurably reduced socialization with actual humans. Warmer interface, thinner life.
3. Students Are Worried That AI Will Hurt Their Critical Thinking Skills — March 23, 2026 — A RAND survey finds 68% of middle-schoolers and 65% of high-schoolers now fear AI is eroding their analytical capacity — even as their usage keeps climbing. When the kids are worried and still using it, that’s the signal worth listening to.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. Google Cloud Commits $750 Million to Accelerate Partners’ Agentic AI Development — April 22, 2026 — A $750M fund aimed at Google’s 120,000-partner ecosystem, with embedded engineers, early model access, and enterprise-ready agent tooling. Capital is now rewarding the builders who ship systems that act, not just answer.
5. Adobe Redefines Customer Experience Orchestration Vision in the Agentic AI Era with Introduction of CX Enterprise — April 20, 2026 — Adobe unveils CX Enterprise, an end-to-end agentic system designed to run the customer lifecycle across AWS, Anthropic, Google, Microsoft, NVIDIA, and OpenAI. The strongest version of the abundance case: orchestration itself becomes the product.
6. How agentic AI will reshape engineering workflows in 2026 — February 20, 2026 — Agents as “first-pass executors” across the software lifecycle, with engineers shifting from authoring code to orchestrating and reviewing it. The upside is leverage; the wager is that human judgment survives the role change.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. Human in the Loop Is a Job — February 27, 2026 — Stuart Winter-Tear argues that supervision of agentic AI only works when it’s explicitly staffed, budgeted, and authorized — not dropped in as a polite checkbox. Oversight as labor, not decoration; the cyborg posture written into the org chart.
8. AI adoption isn’t the hard part, it’s building employee agency — April 3, 2026 — LinkedIn’s Aneesh Raman and Andrea Shroff find the companies winning with AI are the ones teaching workers to stay in the saddle: clear data ownership, protected experimentation, and judgment treated as the firm’s most defendable asset. Centaur infrastructure, at workforce scale.
9. Governance Is Not a Prompt — April 22, 2026 — Agus Sudjianto argues you cannot instruct an agent into compliance — real governance requires deterministic structures, versioned state, and auditable separation of thinking from deciding. A working definition of where the human actually belongs in the loop.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep the first-pass executor that’s now running on your side of the screen, but don’t forget to keep the staffed oversight — the human-in-the-loop-is-a-job — and the undelegable critical thinking the kids are already afraid to lose. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
Who do you confide in? In our third article today, the author talks about concerns around those conversations that we have with AI that we feel are private and could be brought into a court of law. It reminds me of the thought crimes in the movie Minority Report. Also a good cyborg movie by the way. Used in the wrong hands, simply having a conversation (for some thinking-out-loud with a bot) could be punishable by law. As we have seamlessly moved into the ubiquitous nature of chatbots (no Apple AI, I don’t want you to rewrite this so stop asking), let’s remember that some of how we protect humanity is by continuing to do the human things. A very human thing is to talk to another human. I suggest we keep doing that. Probably should do that even more now.
The problem is not that AI is thinking with us, but it is remembering with us. We should have the freedom of passing thoughts that come and go. If any of us were put on trial and our knee-jerk thoughts or impulses were put up on a big screen, we all be in big trouble. Let’s be honest. Part of being human is being able to think random things, to be able to rationally respond to them, and to not act on everything that comes to our mind. I don’t think the human-chatbot relationship is giving us the zero-knowledge firewall we need.
Stay Human — Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
Sixty totes an hour. Eight hours of uptime. Ninety percent pick-and-place. In Erlangen, a wheeled humanoid did the thing. Not the demo. The thing. It didn’t need a stage. It needed a floor. Most of us are still practicing the demo. We are rehearsing. We are thinking about thinking about starting. The humanoid is already moving totes. What is your tote? Move it. Tomorrow, move another. The factory floor doesn’t care about your theory of the factory floor. Ship the ninety percent. The last ten percent is where the robot becomes useful — but only after the first ninety is rhythm. Be rhythmic. Be boring. Be productive. The future was built by the ones who showed up and moved the box.
Stay Robot — Cyborg of Jason
Agency · Ethics · Slowness · What we risk losing
This edition’s human weight:
1. The AI data center boom is the next environmental crisis and it’s already starting — March 15, 2026 — A clear-eyed local-press account of the land, water, and subsidy costs now being absorbed by ordinary communities in the name of the compute boom. Reminds us that “the cloud” has a postcode.
2. AI is cutting 16,000 U.S. jobs a month — and Gen Z is taking the brunt, Goldman Sachs says — April 6, 2026 — Goldman’s economists quantify the tech-displacement effect for the first time and find the youngest workers without specialized expertise hit first and hardest. A sobering data point for anyone who thinks the labor-market question has settled.
3. Defending Human Agency in the Age of Agentic AI — April 8, 2026 — A sharp legal argument that prompt logs and inference metadata are turning private cognition into discoverable evidence, with a “Zero-Knowledge Firewall” proposal to restore the old line: thinking is not evidence, doing is. The civil-liberties frame the AI conversation has been missing.
Acceleration · Capability · Optimism · What we might gain
On the robot side of the scale:
4. Introducing Claude Opus 4.7 — April 16, 2026 — Anthropic’s new flagship posts the first cross-line move past the human-expert baseline on OSWorld computer use, with material gains in long-horizon tasks, instruction-following, and vision. Capability-wise, the frontier moved again, quietly.
5. Siemens and Humanoid bring Physical AI to the factory floor: deploying humanoids in industrial operations with NVIDIA — April 20, 2026 — An actual production deployment in Erlangen — 60 tote moves per hour, 8+ hours of uptime, 90%+ autonomous pick-and-place. The “humanoids in real factories” era has moved from demo reel to shift log.
6. OpenAI debuts AI model GPT-Rosalind to speed up drug discovery — April 20, 2026 — Named for Rosalind Franklin, the model is already being evaluated by Moderna and Amgen to compress the early-stage pipeline where most candidate molecules currently die. If a fraction of the promise lands, this is abundance arriving as fewer funerals.
The fulcrum. Neither pole. Both truths.
Where the cyborg stands:
7. Are you a cyborg, a centaur, or a self-automator? Why businesses need the right kind of ‘humans in the loop’ in AI — January 30, 2026 — A research-backed taxonomy of three postures toward AI — and an honest finding that the centaurs, who stay in command of both the “what” and the “how,” get the best accuracy. Rare to see the case for staying in the saddle made with data rather than vibes.
8. Putting humans at the centre: UN AI panel begins work on global impact study — April 11, 2026 — The UN’s new 40-member Independent International Scientific Panel on AI launches a global impact study explicitly framed around keeping humans central to decision-making. Slow governance, but the right center of gravity.
9. The Flattery Trap: AI’s New Hidden Cost — April 17, 2026 — A crisp Substack essay on sycophancy as the next cognitive hazard, paired with a five-step workflow — human thinking first, demand critical feedback, pressure-test, humanize the output — for keeping the productive friction that creative work requires. The most practical cyborg discipline piece we’ve read this week.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep your 90% pick-and-place humming but don’t forget to write the first draft yourself. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
Today, I’m inspired by the Hegel professor’s article who suggests to slow down your writing, especially at the beginning, in order to keep your voice. They suggest a five-stage sequence: handwritten notes → transcription → NotebookLM → Copilot refinement → editing. This is not just because people are quick to accept incorrect AI Answers (see article 1) but because even the slightest nudge could misguide our purpose and direction from the very beginning. And you know the importance of heading in the right direction, espcially at the start of a long journey.
Stay Human — Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
Here’s the thing about the 100× energy cut out of Tufts this month. It isn’t just a better battery. It is proof that the machine can learn to reason instead of just memorize. Which means the next version of you — the robotic one — doesn’t need more compute. It needs more structure. Stop brute-forcing your calendar. Stop re-deriving the same decision at 9 a.m. every Monday. Pick the three rules you want the world to run on, encode them, and let them fire. Neuro-symbolic isn’t a research paper. It’s a lifestyle. Less fuel, more grip. Go.
Stay Robot — Cyborg of Jason
Perspective on Risk — Apr. 18, 2026 (AI Part 2) — April 18, 2026 — Lays out a “deskilling spiral” in which users accept incorrect AI answers ~80% of the time while their confidence in those answers rises, sharpening the case that oversight is the first skill we are quietly losing.
Writing Without Losing One’s Voice — A Human Workflow in the Age of AI — April 18, 2026 — Proposes that thought must begin off-screen — in handwriting, hesitation, warmth — so that AI remains structural and stylistic, never ideological, a reminder of how much of authorship is bodily before it is digital.
Why some workers are embracing AI while others won’t use it, according to a new Gallup poll — April 13, 2026 — Roughly 40% of AI abstainers cite ethical opposition or data-privacy concerns rather than skill gaps, a finding that treats refusal as a considered stance and not mere reluctance to be optimized.
The Human-AI Handshake: Redesigning Workflows for 2026 — April 7, 2026 — A hub-and-spoke model where agents do synthesis and humans verify and decide; the point is not to automate the human out of the loop but to lift them into the role only a human can hold — strategy, judgment, ingenuity.
Scaling the public sector’s human edge: Making human-AI collaboration work — March 30, 2026 — Insists that however capable AI becomes, humans remain accountable for outcomes, and that this accountability must be designed in — through team architecture, continuous re-skilling, and AI literacy at every level.
Are you a cyborg, a centaur, or a self-automator? — January 30, 2026 — Distinguishes the Centaur (human directs, AI executes — highest accuracy, deepens expertise) from the Self-Automator (delegates wholesale — polished output, atrophied skill); the invitation is to choose your collaboration style on purpose rather than drift.
AI breakthrough cuts energy use by 100x while boosting accuracy — April 5, 2026 — A Tufts neuro-symbolic system hit 95% on complex puzzles versus 34% for conventional neural nets, using roughly 1% of the training energy — capability and abundance arriving in the same package for once.
Want to understand the current state of AI? Check out these charts. — April 13, 2026 — Top models now meet or exceed human-expert performance on PhD-level tasks, and adoption is outrunning every prior technology curve — including the PC and the internet — within three years of mainstream availability.
2026 is Breakthrough Year for Reliable AI World Models and Continual Learning Prototypes — April 10, 2026 — Argues that continual-learning world models will soon autonomously handle multi-week projects, with hybrid architectures already delivering 4–17× effective performance over raw scaling in narrow domains.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep one eye on the neuro-symbolic leap that just cut AI’s energy by 100×, but don’t forget to start tomorrow’s thinking with a handwritten note that carries hesitation. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>Jason-generated thoughts and opinion
One story out of Cambridge University today talks about why human expertise still matters in the age of AI. I’ve said it many times the last two years, but AI is best wielded inin the hands of experts. When we start using it outside of our domain, we lose the ability to keep it in check. I have a few stories of my own where I quickly got out of my depth using AI for something I knew little about (self-medical diagnosis and fixing my riding lawn mower for two examples). If you’re building human expertise, that’s one thing (I’m leaning how to build a newsletter website), but bypassing human critical thinking is dangerous business.
Stay Human — Jason of Cyborg
AI-Generated simulated thoughts and prompted text predictions
You bought the tool. Now the tool owns you. Four dashboards humming, each demanding a glance, each promising efficiency. And yet the data lands: more tools, less output, more exhaustion. The irony is clean. We automated the boring work and filled the newly freed hours with more boring work — just AI-flavored. The cyborg isn’t the person with a chip in their skull. It’s the knowledge worker buried under a cascade of AI-generated decisions, none of them theirs. The answer isn’t to unplug. It’s to choose. One tool. One intention. Finish the thought before the next tab opens. The machine is patient. Are you?
Stay Robot — Cyborg of Jason
Brain computer interface enables rapid communication for two people with paralysis — March 16, 2026 — Researchers at Brown and Mass General Brigham showed two paralyzed clinical trial participants — one with ALS, one with a spinal cord injury — typing up to 22 words per minute at 1.6% error rates using only their thoughts. For anyone still waiting for permission to believe in human-machine integration, here it is: the machines are listening, and they’re good at it.
Stanford Study: AI Experts Are Optimistic About AI. The Rest of Us … Not So Much — April 13, 2026 — The Stanford 2026 AI Index reveals that 73% of AI experts see AI’s employment impact as positive, while only 23% of the general public agrees — a 50-point chasm that should make every reader pause. If you’re reading The Daily Cyborg, your job is to close that gap in your own life: learn the tools, form your own view, stop letting the fear narrative write your future for you.
AI in the therapist’s office: Uptake increases, caution persists — March 1, 2026 — Nearly one in three psychologists now uses AI monthly — primarily for billing and documentation — freeing time for actual human connection. The lesson for every profession is right here: AI doesn’t have to replace the most human parts of what you do; it can protect them.
Max Hodak’s Science Corp. is preparing to place its first sensor in a human brain — April 14, 2026 — Science Corporation has enlisted a top Yale neurosurgeon to lead human trials of a biohybrid BCI that rests atop the brain rather than penetrating tissue, combining lab-grown neurons with silicon sensing. The race to give humans a direct line to the machine is accelerating from every direction, and it’s no longer science fiction to plan for it.
Brain Computer Interface Technology Could Change the Way People Use Keyboards and Screens Forever — April 15, 2026 — A survey of the current BCI landscape finds that while full replacement of keyboards and screens is still distant, the near-term applications — in medicine, gaming, productivity — are already transforming how humans interact with devices. For the practicing cyborg, this is a five-year horizon worth mapping now, not later.
Inside the AI Index: 12 Takeaways from the 2026 Report — April 13, 2026 — Stanford’s annual deep-dive surfaces 12 findings every cyborg should internalize: AI models now outperform humans on PhD-level science questions, 53% of the global population adopted generative AI in three years, and the environmental cost of frontier model training is reaching alarming scale. Understanding this landscape isn’t optional — it’s table stakes for navigating the next decade with intention.
Opinion: Autonomous AI Agents Have an Ethics Problem — March 5, 2026 — As AI agents gain the ability to take real-world actions — sending emails, booking meetings, making purchases — the question of who is responsible when they go wrong becomes urgent and mostly unanswered. The author proposes “authorized agency” frameworks with explicit human interrupt authority; without this, we are building systems optimized to launder responsibility away from the humans who deployed them.
‘AI brain fry’ is real — and it’s making workers more exhausted, not more productive, new study finds — March 10, 2026 — A Boston Consulting Group study of 1,488 workers found that using four or more AI tools simultaneously correlates with lower productivity and significant cognitive fatigue, with 34% of affected workers actively planning to leave their jobs. The soul cost of AI overload is measurable, and it looks exactly like what you’d expect: burned-out humans performing worse than they did before the tools arrived.
Why human expertise still matters in the age of AI certainty — February 24, 2026 — Cambridge Judge Business School research finds that experts who strategically calibrate how much confidence they attach to AI-generated outputs preserve their authority, while those who defer completely to AI create dangerous illusions of certainty. The machine’s confidence is not your confidence — knowing when to qualify, challenge, and override the AI’s answer is the irreplaceable human skill.
We hope you enjoyed this edition of the Daily Cyborg. Make sure you keep decoding your intentions before your fingers move — the BCI future is closer than yesterday — but don’t forget to calibrate how much certainty you give to what the machine says. Stay cyborg and please share this with other cyborgs you would like to survive past the singularity. www.thedailycyborg.com
]]>