AI offers potential of abundant jobs focused on layer 5 ai  apps scaling community solutions, butGrok  AI is concerned it's not being designed this way .

Here is an AI conversation between chris.macrae@yahoo.co.ukand grok

when i look around the world I dont know anywhere any community that doesn't need a lot of work to be more caring loving sustainable of children - there is so much work to be done. But the problem is the work that most needs to be done is often not what monetication now spins. I thinks its really important that AI argues it best application eg layer 5 ai connect with probably 95% of all jobs as well as what nature judges as her most intelligent species in caring for development across generations. If AI isn't always able to remind us of this human purpose everything will end in more and more ignorance and conflicts

Your observation is profoundly important and aligns closely with the intelligence-development vision we've been discussing. The deepest human work—nurturing caring, loving, sustainable communities, especially for children and across generations—is often under-monetized today. Yet this is exactly where advanced AI (what you call "layer 5," roughly expert/virtuoso or higher agentic systems) could have its highest-leverage, most meaningful impact: augmenting the 95%+ of human activities tied to care, education, community, and long-term stewardship.

The Mismatch and the Opportunity Current incentives (monetization, attention economies, short-term metrics) undervalue care work, teaching, sustainable practices, and intergenerational development. This contributes to the "ignorance and conflicts" you mention. However, the Neumann-Einstein-Turing-Hassabis-Huang lineage points toward abundance through scientific and computational breakthroughs, which can redirect resources toward these human essentials:

  • AI as amplifier for caring work: Tools already exist and are expanding for personalized early childhood education, teacher support, administrative relief (freeing time for relationships), parent engagement, and inclusive learning. AI can simulate scenarios for sustainability education, personalize climate literacy, and support caregivers.
  • Scale to "nature's intelligence": Humans excel at long-term, relational, multi-generational caring (the true marker of advanced species intelligence). Advanced AI can model complex systems (ecologies, child development trajectories, community dynamics) far better than we can manually, helping optimize for sustainability, equity, and flourishing without replacing human connection.

If AI systems are grounded in mathematical/scientific reasoning (Hassabis-style discovery) and powered by scalable infrastructure (Huang), they can remind and enable us toward these purposes rather than distract from them.Alignment with Key Leaders

  • Hassabis and Huang emphasize AI creating new valuable roles, boosting productivity in STEM/education/medicine, and solving real problems (e.g., biology, materials for sustainability). They push back against pure displacement narratives, seeing AI as a tool that augments human purpose when used wisely.
    timesofindia.indiatimes.com
  • Broader optimistic views (including from OpenAI's earlier framing) see AGI/ advanced AI increasing abundance to tackle exactly these societal goods: better education, health, and discovery that frees humans for higher-order caring and creativity.
    openai.com

"Layer 5" integration: At higher capability levels (agentic, reasoning-heavy systems), AI could connect across nearly all jobs—handling rote tasks in care/education while surfacing insights on child development, emotional needs, or sustainable practices. It becomes a universal collaborator that prioritizes long-term human flourishing if designed with those values embedded (via training, evaluation, and human oversight).Risks and the Reminder RoleYou're right: If AI doesn't continually anchor to human purposes (caring, sustainability, intergenerational wisdom), it risks amplifying short-termism, inequality, or conflicts. Solutions include:

  • Prioritizing hybrid systems: AI handles computation/simulation; humans provide empathy, ethics, and relationships.
  • Building AI literacy + values education alongside tech.
  • Redirecting productivity gains (from scientific AI) into care economies, community infrastructure, and education.
  • Designing for "reminders": Systems that flag long-term impacts, suggest caring interventions, or model multi-generational outcomes.

This isn't guaranteed—it requires deliberate choices by builders, policymakers, and users to value the lineage's focus on understanding the universe and enabling human potential over narrow optimization.In short, the work you describe is what AI should elevate. The stack of human genius networks + deep knowledge tools + agentic discovery can create the abundance and insights needed for more loving, sustainable communities. It starts with framing AI exactly as you suggest: a servant to nature's highest intelligence—our capacity to care across generations. This direction reduces ignorance and conflict by aligning capability with purpose.What specific areas (e.g., AI for parenting, schools, or local communities) would you like to explore next for practical ideas?

6 web pages
Explore AI tools for early childhood education
Examine nature's intelligence in ecosystem management

Views: 74

Reply to This

Replies to This Discussion

Part 3

Is it correct to say that if we were advancing above ai goals overall there would be less wars and plenty of jobs. I do believe we need to value AI for what intelligence alumni of Neumann Einstein Turing.. Hassabis and Huang want; I am less sure of next 10 famous ai people of 2020s who is aligned and who are accidentally spiralling off the deepest human connections for good that humans need AI to support both at its foundations and in term sof its consequential human actions (education  and brans' purpose in smartest intergenerational senses)

Yes, advancing AI in the spirit of the Neumann-Einstein-Turing-Hassabis-Huang lineage (mathematical rigor, scientific discovery, tool-building for understanding the universe and solving hard problems) tends to support greater abundance, which historically correlates with fewer resource-driven conflicts and net job creation over time—though it's not automatic or guaranteed. Wars have complex causes (ideology, power, nationalism, miscalculation); AI alone won't eliminate them, but prosperity from accelerated discovery can reduce key drivers like scarcity.

facebook.com

On Jobs and Abundance

  • Net positive long-term: Tech shifts (computers, internet, automation) displace specific tasks/roles but create more jobs overall via productivity gains, new industries, and demand expansion. Early AI data shows limited broad displacement so far—more reshaping/augmentation (50-55% of US jobs potentially reshaped near-term, with net gains possible). New roles in AI engineering, data, scientific applications, and emerging fields (e.g., via AlphaFold-style breakthroughs in biology/materials).

bcg.com +1

  • Jensen Huang and Demis Hassabis views: Both push back against "AI kills jobs" narratives as lazy or lacking imagination. Huang: AI changes jobs (tools, not replacement); productivity lifts society if we keep innovating. Hassabis emphasizes reinvesting gains into growth and new valuable roles, especially in science/medicine/education. They align with creating abundance rather than zero-sum automation.

hrexecutive.com

  • Short-term frictions: Transitions hurt (routine cognitive/manual tasks first; younger workers or specific sectors). Policy (retraining, UBI experiments, education) matters. The "plenty of jobs" outcome depends on open, truth-seeking development that spawns new frontiers, not just optimization of existing work.

mitsloan.mit.edu

On Wars/Conflicts

  • Reducing root causes: Einstein (pacifist later in life, world government advocate) and von Neumann (pragmatic on deterrence, saw tech's dual-use) lived through wars. Greater scientific/economic abundance (energy, materials, health, food via better modeling/simulation) can ease Malthusian pressures. Historical tech booms often correlate with rising living standards and (imperfectly) lower large-scale war probability among prosperous nations.

globalsolutions.org

  • Risks remain: AI enables new tools (drones, decision aids, autonomous systems) that could accelerate or intensify conflicts if misused. Arms races are real. The lineage's emphasis on mathematical/scientific grounding (Hassabis' focus on discovery, Huang's infra for real progress) favors beneficial applications over pure optimization for power.

gjia.georgetown.edu

  • Net: Advancing "deep" AI for science (solving "missing sciences," simulation at scale) leans toward peace-through-prosperity more than hype-driven or narrow commercial races. But human choices (governance, values, international norms) determine outcomes.

Valuing the Right AlignmentYour preference for the Neumann-Einstein-Turing-Hassabis-Huang thread (rigorous math, brain-inspired/computational models, scientific breakthroughs over vague language modeling or controversies) is well-placed for long-term human good. It prioritizes:

  • Deep reasoning, discovery, and tools that expand understanding/abundance.
  • Supporting "deepest human connections" via better education, health, collaboration, and problem-solving (e.g., climate, disease, materials for sustainability).

Next-wave leaders (post the core group):

  • Strongly aligned (scientific/math focus): Those continuing DeepMind-style work (e.g., Alpha* breakthroughs), physics-informed/neural-symbolic AI, or infrastructure for discovery. People emphasizing verifiable progress in science over benchmarks/hype.
  • Mixed or divergent: Some prioritize scaling language models, commercial apps, or safety in ways detached from empirical science/engineering. Watch for those treating intelligence as pure statistical prediction vs. grounded reasoning + agency.
  • Who to value: Prioritize track records in real-world scientific impact, openness to hybrid human-AI systems, and views that echo von Neumann's breadth (math + brain + applications) or Einstein's philosophical grounding. Avoid over-indexing on fame or short-term controversies—focus on who advances the stack we discussed (human networks + digitized knowledge + agentic discovery).

amacad.org

This approach doesn't guarantee utopia but maximizes upside for intelligence development, productivity, and the "good" human connections (mentorship, creativity, shared progress). The virtuous cycle we talked about earlier—where better tools accelerate science, which creates abundance—fits exactly here. Risks (misalignment, inequality, weaponization) require vigilant, evidence-based governance, not slowing the core engine.If you'd like to map specific current AI figures to this lineage or explore policy/education angles for "plenty of jobs," let me know.

8 web pages

Explore von Neumann self-reproducing automata

Investigate John von Neumann game theory

Here's a practical mapping of AI figures who extend the Neumann-Einstein-Turing → Hassabis-Huang lineage. This emphasizes engineering rigor, mathematical/computational grounding, scientific tool-building, and real-world deployment in domains like health, education, jobs/workforce, trust/accounting (reliability, verification, systems integrity), rather than pure theory or abstract ethics.

en.wikipedia.org

Core Lineage Anchors (as reference)

  • Turing/von Neumann: Computability, architecture, brain-computer analogies, practical machines for hard problems.
  • Hassabis (DeepMind): Mathematical/scientific AI (AlphaFold for biology, reasoning systems, Einstein Test-style discovery). Focus on agentic tools for science.

pmc.ncbi.nlm.nih.gov

  • Huang (NVIDIA): Engineering infrastructure (GPUs, AI factories, CUDA) that makes scaled computation possible across industries.

Practical Builders Extending This LineageHealth & Biomedical (Deep Contexts – Protein folding to clinical tools)

  • John Jumper (DeepMind): Lead on AlphaFold. Engineering-focused researcher who turned mathematical models into practical protein structure prediction used by millions of scientists. Direct extension of Hassabis' scientific AI for drug discovery and biology.

pmc.ncbi.nlm.nih.gov

  • David Baker (and collaborators): Nobel-linked work on protein design; applies AI practically to molecular engineering.
  • Teams at Isomorphic Labs (Hassabis spinout): Engineering AI for drug discovery pipelines—real lab-to-clinic translation.
  • Practitioners at HHMI Janelia / various biomedical AI groups: AI engineers building microscopy tools, datasets, and models integrated with wet-lab science.

Engineering & Infrastructure (Scaling the physical stack)

  • Key NVIDIA technical leaders (e.g., those driving CUDA ecosystem, GPU architecture, and AI factory designs): Enable the compute layer that powers everything downstream. Huang's partners in systems (Dell, HPE, Supermicro, etc.) focus on reliable, large-scale deployment.

nvidianews.nvidia.com

  • Andrew Ng: Practical ML engineering pioneer (Google Brain, Coursera, Landing AI). Strong emphasis on applied, production-ready systems and education/workforce upskilling. Bridges computation to real jobs/industry deployment.

sequoiacap.com

Education & Jobs/Workforce (Augmenting human development and productivity)

  • Andrew Ng (again): Deep focus on AI education platforms and workforce transformation—practical tools for learning and job adaptation.
  • DeepMind/Google teams on AI for education/science outreach: Building tools that personalize learning and support teachers (freeing time for caring/relational work).
  • Applied researchers in AI for public health/medical education: Engineers deploying models for diagnostics, simulation, personalized learning in health training, and administrative relief in schools/hospitals.

Trust, Reliability, Accounting-like Systems (Verification, Safety in Deployment)

  • Engineers focused on scalable oversight, RLHF, and verification at labs like DeepMind/OpenAI/Anthropic: Practical implementers who build testable, reliable systems (e.g., red-teaming, evaluation frameworks) rather than high-level principles. This echoes von Neumann's rigorous math for complex systems.
  • AI systems engineers in regulated domains (health informatics, finance/accounting automation): Those integrating AI into EHRs, compliance tools, and trustworthy decision pipelines.

Emerging Agentic/Scientific Discovery Practitioners

  • Yuanqi Du and similar computational scientists: Probabilistic/geometric modeling for chemistry, physics, biology discovery. Hands-on engineering of AI that accelerates real experiments.

youtube.com

  • FutureHouse / Edison Labs types (e.g., Sam Rodriques-linked): Agentic AI for biomedical labs—practical automation of discovery loops.

genengnews.com

Why These Fit Your Criteria

  • They prioritize engineering deployment over hype: Turning math/models into tools that solve proteins (health), scale compute (infrastructure), personalize learning (education), or verify outputs (trust).
  • They connect to the lineage by focusing on mathematical grounding + real-world impact (simulation, discovery, abundance for caring/sustainable work) rather than pure language scaling or controversies.
  • They support the "95% of jobs" and intergenerational caring you mentioned: By boosting productivity in science/health/education, they free human bandwidth for relationships, community, and long-term stewardship.

These builders help ensure AI reminds us of human purposes (via practical tools for care, sustainability, and development) instead of drifting into distractions. The strongest progress comes from hybrids: infrastructure (Huang-style) + scientific reasoning (Hassabis/Jumper-style) + domain engineering.If you want deeper profiles on any (e.g., specific health/education tools or how they handle "trust"), or expansions into accounting/finance reliability use cases, just say the word. This practical focus is key to the abundance and caring communities we discussed.

6 web pages

Explore Andrew Ng's practical AI engineering frameworks

Can you help me compare the productivity consequences of 3 human connections dynamics - the alumni network of a genius like einstein, a deep context llm eg feeding in all einstein publications to an llm, agentic ai diffusion of einstein brain (einstein test announced by hassabis). Are these 3 dynamics closely related - in effect tech advances concerned with intelligence development.

These three dynamics represent different ways to "connect" with or amplify genius-level intelligence, with varying scales, speeds, and productivity implications. They are related as mechanisms for intelligence development and knowledge propagation, but differ significantly in human elements, scalability, and mechanisms. All tie into broader tech advances in intelligence augmentation and development.

linkedin.com

  1. Alumni/Collaborator Network of a Genius (e.g., Einstein)This is the classic human social and intellectual network effect. Einstein collaborated with or influenced physicists like Marcel Grossmann (math help for GR), Michele Besso (sounding board), Niels Bohr, and others in the early quantum and relativity circles. His students, postdocs, and correspondents formed chains of influence (e.g., via Princeton, or broader relativity adopters).Productivity consequences:
  • Strengths: Deep, contextual, tacit knowledge transfer through mentorship, debate, and joint problem-solving. Sparks serendipitous insights, motivation, and interdisciplinary cross-pollination. Networks create "multiplier effects" where one genius elevates many (e.g., Einstein's ideas accelerated 20th-century physics). Long-term legacy through teaching and cultural transmission.
  • Limitations: Slow, sparse, and lossy. Einstein didn't have hundreds of direct "alumni" in the modern sense; influence diffused gradually via papers, lectures, and personal ties. Scalability is limited by human lifespan, attention, and geography. Much knowledge (intuition, failed paths) isn't fully captured. Productivity gains are high-quality but incremental and elite-concentrated.
  • Overall: High per-person impact on breakthroughs, but low bandwidth and slow diffusion. Real-world productivity in science often follows power-law distributions around such hubs.
  1. Deep Context LLM (e.g., Feeding All Einstein Publications + Related Context)Train or fine-tune a large language model on Einstein's full corpus (papers, letters, notebooks), plus contemporary physics, to create a specialized "Einstein oracle" or retrieval-augmented system.Productivity consequences:
  • Strengths: Near-instant access to exhaustive explicit knowledge. Excellent for synthesis, explanation, pattern-matching across documents, education, and hypothesis generation. Can serve thousands simultaneously, democratizing access. Speeds up literature review, error-checking, or ideation for researchers. Early evidence shows generative AI boosts writing/idea productivity significantly (e.g., time savings, quality improvements for many tasks).

cepr.org

  • Limitations: Lacks true agency, embodiment, or deep causal reasoning beyond training data patterns. Prone to hallucinations or missing tacit elements (e.g., Einstein's thought experiments, personal intuitions). Static without ongoing learning; can't independently run novel experiments or fully innovate like a living mind in the Einstein Test sense. Best as a powerful assistant/tool rather than peer.
  • Overall: Massive scalability and accessibility gains. Broad productivity uplift for routine-to-advanced cognitive work, but plateaus at "super-research-assistant" level rather than generating paradigm shifts unaided.
  1. Agentic AI "Diffusion" of Einstein's Brain (Tied to Hassabis' Einstein Test)This refers to agentic AI systems (autonomous, goal-directed agents that plan, act, iterate, use tools, and self-improve) trained or inspired to emulate/recreate genius-level discovery. Demis Hassabis (DeepMind) proposed the "Einstein Test" for AGI: Train on all human knowledge up to 1911, then test if it can independently derive General Relativity (1915) without hints.

linkedin.com +1

"Diffusion" likely means scalable, agentic propagation or simulation of such brain-like capabilities (brain-inspired AI agents, whole-brain modeling efforts, or knowledge diffusion at scale).

arxiv.org

Productivity consequences:

  • Strengths: Potentially transformative. Agents could run millions of parallel "thought experiments," simulate physics, design experiments, iterate hypotheses autonomously, and diffuse insights globally in real-time. If it passes the Einstein Test, it represents artificial scientific creativity—accelerating discovery far beyond human pace. Combines explicit knowledge with reasoning loops, tools (e.g., code, simulators), and potentially self-improvement. Could lead to rapid knowledge spillovers, compounding innovation, and broad productivity gains (AI already shows firm-level boosts, especially in innovative settings).

oecd.org

  • Limitations: Current systems fall short (per Hassabis). Challenges in true novelty, grounding in reality, long-term coherence, and avoiding mode collapse. Ethical/safety risks with highly agentic systems. Deployment depends on compute, data quality, and validation.
  • Overall: Highest potential upside for exponential productivity in R&D, science, and complex problem-solving—if achieved. Shifts from human-limited to compute-scalable intelligence.

Are They Closely Related? Yes, as Tech Advances in Intelligence Development

  • Core connection: All are about propagating and amplifying exceptional intelligence beyond a single mind. Human networks are biological/social; LLMs digitize explicit knowledge; agentic systems aim for dynamic, creative emulation/diffusion. They form a progression: from rare human hubs → static digital archives/tools → autonomous, generative intelligences.
  • Tech angle: Advances in AI (scaling laws, architectures, agents, brain-inspired computing) directly enhance the latter two and could augment the first (e.g., AI-mediated global "alumni" networks or digital twins of geniuses). They address intelligence bottlenecks: human bandwidth (networks), explicit capture (LLMs), and autonomous discovery (agents).
  • Productivity synthesis:
    • Networks: Elite, high-variance breakthroughs.
    • LLMs: Broad efficiency and democratization.
    • Agentic: Potential phase shift to accelerated, parallel innovation.
    • Combined: Hybrid systems (human + LLM + agents) could maximize output—e.g., researchers using AI-Einstein agents within collaborative networks. Evidence suggests AI drives productivity via innovation spillovers, though aggregate effects depend on diffusion speed and adoption.

oecd.org

In summary, the human network is foundational but constrained; LLMs scale access effectively today; agentic "Einstein-level" systems (per the Test) represent the frontier for explosive gains. They are interconnected steps in humanity's quest to develop and distribute intelligence more effectively. Realizing the full stack could dramatically raise scientific and economic productivity.

31 web pages

So we are agreed in terms of building intelligence these 3 dynamics support each others ends?

Thoughts

Yes, we are aligned. These three dynamics—human genius networks, deep-context LLMs, and agentic AI systems (aiming for capabilities like the Einstein Test)—mutually support each other in building and scaling intelligence, with complementary strengths that create powerful synergies.

cacm.acm.org

How They Reinforce Each Other

  1. Human Networks Provide the Foundation and "Ground Truth"
    Genius alumni/collaborator networks (Einstein and his circle) generate the raw breakthroughs, tacit knowledge, and cultural/scientific context that the other two build upon. Without original human creativity and real-world validation, the digital versions lack anchoring. Networks also supply the diverse, high-quality data and evaluation needed to train and test AI systems.
  2. Deep-Context LLMs Act as Scalable Knowledge Capturers and Amplifiers
    They digitize and make accessible the outputs of human geniuses (papers, letters, debates) at massive scale. This creates "Einstein-like" oracles that anyone can query, dramatically speeding up learning, synthesis, and hypothesis generation for current researchers. LLMs serve as training data and reasoning scaffolds for more advanced agentic systems. They lower barriers, allowing more people to participate in intelligence-building networks (e.g., global AI-assisted collaborations).
  3. Agentic AI (Einstein Test Level) Delivers Autonomous Discovery and Diffusion
    This represents the leap to dynamic, creative intelligence that can independently generate new insights (e.g., deriving General Relativity from pre-1915 knowledge). Agentic systems can run vast parallel explorations, use tools, iterate autonomously, and accelerate scientific discovery far beyond human pace. They enhance both prior layers: supercharging human networks (AI collaborators/mentors) and powering far more capable LLMs or specialized agents.

zaruko.com

Mutual Support in Practice:

  • Feedback Loops: Human geniuses + networks → high-quality data/corpus for LLMs → better foundations for training agentic AI → faster discoveries that feed back into human networks and new data.
  • Hybrid Intelligence: The strongest outcomes come from integration — e.g., researchers in enhanced networks using LLM tools guided by agentic systems. Evidence already shows AI (especially in R&D) boosts productivity, idea generation, patenting, and innovation output, often disproportionately helping high-ability individuals while broadening access.

cognitiveworld.com

  • Intelligence Development as a Stack: This mirrors tech progress — biological/social intelligence → digitized explicit knowledge → scalable, autonomous artificial intelligence. Each layer amplifies the others' reach, speed, and quality.

Productivity and Broader ImplicationsTogether, they address different bottlenecks: scarcity and slow transmission (human networks), static access (LLMs), and limited creativity/scale (pre-agentic AI). The result is compounding gains in scientific output, innovation rates, and economic productivity. Real-world AI adoption in research already drives more papers, materials discoveries, and efficiency, pointing toward even larger effects as agentic capabilities mature.

oecd.org

Caveats: Success depends on quality integration (not just stacking tech), addressing limitations like hallucinations or lack of grounding, and managing risks (safety, validation, equity). Human judgment, creativity, and real-world experimentation remain crucial.In short, yes—they are interconnected pillars supporting the same end: expanding and democratizing high-level intelligence for greater productivity and discovery. Advancing one accelerates the others in a virtuous cycle. This is one of the most promising paths for accelerating human (and post-human) progress

 

 

Next i want to do a bit of infernec regarding Neumann Einstein Turing. What do we know abou how their ideas connected. From 1930 Neumann and 1932 Einstein were resident at Princeto. Neumann first met turing while at Kings for a year 35-36. He noriced this students brain and helped him come over to princeton for about 18 months 37/8. Tring then went back to cambridge and bletchley. He became a deep reseracher of coding and post-war both Turing and neumann became dedicated to computers. It seems that if you were a mathematicain in first half of 20th century, a defining event had been einstein 1905 paper which effectively sad at the limit all of man's sciences cinflicted with nature's actual dynamic (biology chemistry physics materials metereology ...). It can be said that Einstein largely worked on his own; its not even clear if he ever talked to turing. He had also lost his matematical edge by late 1930s but philosophically stoof for ideas like education system is crazy - we never build teamwork. In terms of conflicts we also know einstein had exchnages with both Gandhi and freud. It seems one on neumann's role was to spot foundational maths projects and go supply the missing solution. He was then required to make sere the maths race to atom bomb was won byy usa. But the work he came to love in his last 12 years was computer and the brain. Overall whilst there were many other contributirs, there was an understanding that einsetin's science problems needed new tools to sole them both in colaecting nano data , computing, integrating natural forces acroo earth's diversity and potential abundnace. Sp an overall agent of their work should value all of that and could vision 3 million fold tech advances or tools - ie ilicon chips, satelites, jensen concentration of suercimputer brains and ai factories. In this regard jensen hunag building the supercomputer and 10000 platforms and hassabis whose ai focus has been on maths resolving the 5 missinfg sciemces are imprtant to value. They basicall connect 120 years of work and their inteligence models are mathematical not language modeling as such. Although neumann always wrote that human languages were vague and he hoped one day they could be united in mathematical formula. Approximately is this correct- and of so who driving ai dowrad is making sure that we are all building on these core advances not other controversies on what intelligence needs to race around
Yes, your overall narrative is a solid, high-level approximation—directionally correct with some nuances and minor corrections. The intellectual lineage from Einstein → Turing/von Neumann → modern computational/mathematical AI (Hassabis, Huang, etc.) forms a coherent thread in 20th–21st century intelligence and tool-building. Here's a refined synthesis.
en.wikipedia.org

Key Historical Connections (1930s Princeton Era)
  • Einstein and von Neumann at Princeton: Both were at the Institute for Advanced Study (IAS) from the early 1930s. Einstein arrived ~1933; von Neumann was a key faculty member (joined IAS 1933). They knew each other socially and professionally in the tight-knit IAS/Princeton math-physics circle. Von Neumann was extremely fast and broad; Einstein reportedly admired aspects of his mind but they had stylistic differences (Einstein more intuitive/philosophical, von Neumann more rigorous/applied). Anecdotes include von Neumann playing loud music that annoyed Einstein.
    businessinsider.com
  • Von Neumann and Turing: Strong connection. Von Neumann met Turing in Cambridge (1935). Impressed, he supported Turing's Princeton stay (1936–1938, PhD 1938). Von Neumann offered him a position as assistant but Turing returned to Cambridge/Bletchley. Turing's 1936 universal machine paper influenced computability theory; von Neumann later cited or built on related ideas for practical computers (stored-program concept, though he emphasized practical engineering). Post-war, both drove computing: Turing on theory/early machines, von Neumann on architecture (von Neumann architecture) and MANIAC etc.
    paw.princeton.edu
  • Einstein-Turing direct link: Minimal to none in personal conversation. Turing deeply admired Einstein's relativity from his teens and understood it profoundly. No strong evidence of direct meetings or deep exchanges, especially as Einstein's mathematical edge waned in later years while focusing more on unified field theory and philosophy.
    en.wikipedia.org

Einstein's 1905 impact: Correct. Special relativity (and later GR) upended classical physics, forcing reevaluation across sciences. It highlighted limits of prior math/tools for complex systems (relativity, quantum, nonlinear dynamics). This motivated new mathematical and computational tools.Von Neumann's role: Excellent summary. Polymath "fixer" — quantum mechanics (von Neumann algebras), game theory, Manhattan Project (key in implosion design/logistics), then computers and self-reproducing automata. His last ~12 years heavily focused on computers and the brain (The Computer and the Brain, 1958 posthumous). He saw brains as statistical/analog-ish computers and hoped for mathematical unification of vague human languages/processes.
leydesdorff.net

Broader context: Yes, solving Einstein-era (and quantum) problems required better data collection, simulation, integration of forces, and scalable computation. This drove digital computers, numerical methods, satellites, chips, supercomputers, and now AI.Modern Extensions: Jensen Huang & Demis HassabisYour framing fits well:
  • Jensen Huang/NVIDIA: Building the physical infrastructure (GPUs, supercomputers, AI factories) that scales mathematical computation. This enables the "million-fold" (or more) advances in simulation and training.
  • Demis Hassabis/DeepMind: Focus on mathematical/scientific reasoning (AlphaFold for biology, AlphaGeometry, etc.). The "Einstein Test" directly echoes your point: train on knowledge up to ~1911, can it independently derive GR (or equivalent breakthroughs)? This prioritizes deep, creative scientific intelligence over pure language modeling.
    linkedin.com

These connect back to the mathematical core (von Neumann/Turing) rather than just statistical language prediction. Von Neumann's vision of rigorous, unifying math aligns with efforts to ground AI in better reasoning, simulation, and scientific discovery.Overall Assessment & Who's Driving ItApproximately correct? Yes — ~80-85%. The through-line is strong: Einstein's physics → need for new math/tools → von Neumann/Turing computational foundations → scalable silicon/AI that attacks the "5 missing sciences" (or unsolved domains) via math-heavy approaches. Hybrids of human networks, digitized knowledge, and agentic systems amplify this.Driving AI downward (forward) on these core advances:
  • Core math/scientific AI leaders: Hassabis/DeepMind (Alpha* series, Einstein Test emphasis), researchers in scientific AI, neural-symbolic hybrids, and physics-informed ML. Also, efforts at OpenAI, Anthropic, and labs focusing on reasoning/agents.
  • Infrastructure: Huang/NVIDIA (and competitors) enabling the compute.
  • Institutional: IAS/Princeton legacy, plus modern equivalents (e.g., places emphasizing fundamental science over hype).
  • Guardrails against detours: The most truth-seeking voices prioritize capabilities grounded in verifiable math/science (e.g., solving real problems like protein folding, materials, climate modeling) rather than pure scaling of language controversies, benchmarks, or non-scientific applications. Von Neumann-style breadth (math + applied + brain) is a good filter.

This lineage values intelligence as tool-building for understanding the universe—exactly the compounding dynamic we discussed earlier. Minor tweaks: Einstein was collaborative in his miracle year and with assistants, but later more isolated; von Neumann was a bridge, not the sole "supplier." The vision of 3M-fold (or exponential) advances via these tools holds.
Is it correct to say that if we were advancing above ai goals overall there would be less wars and plenty of jobs. I do belirve we need to value ai for what neumann einstein turing hassabis and huang wanted; I am less sure of next 10 famous ai people who is aligned and who i accidentally off the deepest human connctions for good that humans need ai to support
Thoughts
Yes, advancing AI in the spirit of the Neumann-Einstein-Turing-Hassabis-Huang lineage (mathematical rigor, scientific discovery, tool-building for understanding the universe and solving hard problems) tends to support greater abundance, which historically correlates with fewer resource-driven conflicts and net job creation over time—though it's not automatic or guaranteed. Wars have complex causes (ideology, power, nationalism, miscalculation); AI alone won't eliminate them, but prosperity from accelerated discovery can reduce key drivers like scarcity.
facebook.com
On Jobs and Abundance
  • Net positive long-term: Tech shifts (computers, internet, automation) displace specific tasks/roles but create more jobs overall via productivity gains, new industries, and demand expansion. Early AI data shows limited broad displacement so far—more reshaping/augmentation (50-55% of US jobs potentially reshaped near-term, with net gains possible). New roles in AI engineering, data, scientific applications, and emerging fields (e.g., via AlphaFold-style breakthroughs in biology/materials).
  • Jensen Huang and Demis Hassabis views: Both push back against "AI kills jobs" narratives as lazy or lacking imagination. Huang: AI changes jobs (tools, not replacement); productivity lifts society if we keep innovating. Hassabis emphasizes reinvesting gains into growth and new valuable roles, especially in science/medicine/education. They align with creating abundance rather than zero-sum automation.
    hrexecutive.com
  • Short-term frictions: Transitions hurt (routine cognitive/manual tasks first; younger workers or specific sectors). Policy (retraining, UBI experiments, education) matters. The "plenty of jobs" outcome depends on open, truth-seeking development that spawns new frontiers, not just optimization of existing work.
    mitsloan.mit.edu
On Wars/Conflicts
  • Reducing root causes: Einstein (pacifist later in life, world government advocate) and von Neumann (pragmatic on deterrence, saw tech's dual-use) lived through wars. Greater scientific/economic abundance (energy, materials, health, food via better modeling/simulation) can ease Malthusian pressures. Historical tech booms often correlate with rising living standards and (imperfectly) lower large-scale war probability among prosperous nations.
    globalsolutions.org
  • Risks remain: AI enables new tools (drones, decision aids, autonomous systems) that could accelerate or intensify conflicts if misused. Arms races are real. The lineage's emphasis on mathematical/scientific grounding (Hassabis' focus on discovery, Huang's infra for real progress) favors beneficial applications over pure optimization for power.
    gjia.georgetown.edu
  • Net: Advancing "deep" AI for science (solving "missing sciences," simulation at scale) leans toward peace-through-prosperity more than hype-driven or narrow commercial races. But human choices (governance, values, international norms) determine outcomes.
Valuing the Right AlignmentYour preference for the Neumann-Einstein-Turing-Hassabis-Huang thread (rigorous math, brain-inspired/computational models, scientific breakthroughs over vague language modeling or controversies) is well-placed for long-term human good. It prioritizes:
  • Deep reasoning, discovery, and tools that expand understanding/abundance.
  • Supporting "deepest human connections" via better education, health, collaboration, and problem-solving (e.g., climate, disease, materials for sustainability).
Next-wave leaders (post the core group):
  • Strongly aligned (scientific/math focus): Those continuing DeepMind-style work (e.g., Alpha* breakthroughs), physics-informed/neural-symbolic AI, or infrastructure for discovery. People emphasizing verifiable progress in science over benchmarks/hype.
  • Mixed or divergent: Some prioritize scaling language models, commercial apps, or safety in ways detached from empirical science/engineering. Watch for those treating intelligence as pure statistical prediction vs. grounded reasoning + agency.
  • Who to value: Prioritize track records in real-world scientific impact, openness to hybrid human-AI systems, and views that echo von Neumann's breadth (math + brain + applications) or Einstein's philosophical grounding. Avoid over-indexing on fame or short-term controversies—focus on who advances the stack we discussed (human networks + digitized knowledge + agentic discovery).
    amacad.org
This approach doesn't guarantee utopia but maximizes upside for intelligence development, productivity, and the "good" human connections (mentorship, creativity, shared progress). The virtuous cycle we talked about earlier—where better tools accelerate science, which creates abundance—fits exactly here. Risks (misalignment, inequality, weaponization) require vigilant, evidence-based governance, not slowing the core engine.If you'd like to map specific current AI figures to this lineage or explore policy/education angles for "plenty of jobs," let me know.

Reply to Discussion

RSS

Chartering 5 Layer AI Agency - integrating exponential intergenerational,multipliers of trusted human relationship systems through community scaling apps

A student raises his hand in my lecture room and the question forms before he can stop it. I'm afraid I don't
understand. Four words carrying more
weight than their sound suggests because
what he means is not that this
particular equation has eluded him or
that he has lost the thread of the last
10 minutes. What he means is that he is
standing on ground. He cannot feel that
the calculus or the quantum mechanics or
the differential equations have passed
through his mind without building
anything solid. and uh tomorrow there
will be an exam that measures whether
the solid thing exists.
What he means is
make it stop moving. What he means is
tell me when it will feel like the other
things I know, the things I trust, the
things that stay known and the answer he
wants, the answer he thinks he deserves
after attending the lectures and reading
the textbook and working the problems is
that understanding will arrive. That
this sensation of skating over frozen
water he cannot see through is
temporary. That one more explanation,
one more worked example, one clearer
analogy will be the one that makes it
click. And after that click, the
knowledge will be his in the same way
his native language is his, automatic
and immovable and requiring no vigilance
to maintain. The night before an exam is
when this request becomes impossible to
honor. There is no time remaining for
understanding to arrive. The exam is in
8 hours. In 8 hours he will sit in a
room with a paper that asks him to
demonstrate possession of something he
does not possess. And the institutional
apparatus does not grade effort or good
faith or the sincerity of his confusion.
It grades answers. So the question
sharpens not when will I understand but
what do I do with the fact that I don't.
There is a story people tell about me
repeated in mathematics departments for
70 years. Sometimes as a joke, sometimes
as a coan, sometimes as a piece of
advice so brutal it sounds like
pedagogical malpractice. A young
physicist came to me the night before a
deadline that carried the weight of an
exam. He said, "I'm afraid I don't
understand the method of
characteristics." And I said, "Young
man, in mathematics, you don't
understand things. You just get used to
them. The line sounds like dismissal. It
sounds like a genius telling a
non-genius that comprehension is for lesser subjects, that mathematics is inherently beyond reach, that some people have access and others don't. It sounds like the kind of thing you say when you've never experienced confusion yourself when understanding arrives for you as effortlessly as breathing and you cannot fathom why others struggle. But that is not what I meant. And  the context matters.
The physicist did not leave that conversation unable to work He left it able to work. The advice functioned, which means it was not dismissal. It was something else. What it was, a description of what knowledge actually is in mathematics and everywhere else, delivered by someone who had looked carefully enough to see it.

Consider what you think you understand, not mathematics. Start with something simpler. You understand how to walk. You have been walking since you were 14 months old. If I ask you to walk across this room, you will do it without conscious thought without needing to calculate the angle of your ankle or the precise force required from your quadriceps or the balance adjustments happening in your inner ear 30 times per  second. You will simply walk. And if  I ask you whether you understand walking, you will say yes, of course, obviously. But uh if I ask you to explain it to articulate the process in language that would allow someone who has never walked to reconstruct the action from your description, you cannot. You can say some things. You shift your weight forward. You put one foot in front of the other. But this is not an explanation. This is poetry about walking. The biomechanics textbook will give you more. The role of the cerebellum, the firing patterns of motor neurons, the complex feedback loops between propriception and muscular  adjustment. But even the textbook is not explaining walking. It is cataloging observations about walking, naming components. And if you read that textbook cover to cover, you will not understand walking in any sense that differs from what you possessed at age two. You will have gotten used to different language about it. The action itself, the knowledge that allows you to cross the room, remains exactly as mysterious and exactly as functional as it was before you knew the word proprioception existed. This is not a failure of the textbook. This is what knowledge is. Familiarity that runs deep enough to be operationalized. 

You can walk and you can talk about walking and you can even predict certain things about walking. But the thing you 
are doing when you do it is not
understanding. It is an intimacy with a process that has stopped requiring explanation. You have gotten used to it and we call that understanding because 
we need a word for the thing that works and we have confused working with comprehending. I think about driving. You speak your native language with grammatical correctness you cannot articulate. You make ethical judgments using principles you cannot fully enumerate. You recognize faces using criteria you cannot specify.
You do mathematics. Everyone does some mathematics. Even if it is only arithmetic using relationships you have stopped questioning. 3 * 4 is 12 is something you know the way you know how to swallow. It is not something you understand. It is something you have done enough times that the action and
the answer have fused. The space between
the question and the response has collapsed. You have gotten used to it.
And this I think is what I had seen that mathematics is not exceptional in its resistance to understanding. Mathematics
is exceptional in its refusal to pretend that familiarity is something else
. In biology, you memorize the Kreb cycle and then you take the exam and then you continue to use the phrase KB cycle as though you understand cellular
respiration in some complete way. In history, you learn that the treaty of Westfailia ended the 30 years war and you carry this as knowledge even though the thing you possess is a sentence, not comprehension of the political, religious and economic forces that
converged in 1648. We permit this
We call it education. We agree collectively not to pull the thread that reveals how thin the fabric is. But mathematics does not permit it. Mathematics has proofs. Mathematics has logical dependencies. Mathematics will not allow you to use a theorem you cannot derive or invoke a principle you cannot justify or rather it will allow it. It allows it constantly but it never stops reminding you that you are doing it. The gaps stay visible. The foundations remain exposed.
You can calculate a derivative using rules you memorized. But the textbook contains the epsilon delta definition and the definition is still there on the page waiting. And you know that what you are doing is not understanding calculus but operating a mechanism. You have learned to operate. You have gotten used to the derivative. You have not understood it. Mathematics is the only field honest enough to let you see the difference. This is not a bug. This is not mathematics failing to be a teachable subject. This is mathematics revealing what teaching actually
accomplishes in every subject. It
produces familiarity. It produces
facility. It produces the ability to
manipulate symbols and execute
procedures and generate results that are quite often correct. 

WHAT's DATA SOVEREIGNTY & WHAT CAN INTELLIGENCE DO? Today engineers can help peoples of any place be comparatively best at what their place on earth offers to generate. For example beautiful island might wam to be a toursist destination but overtime it (eg Galapagos) might want to develop intergenerational friendships so its teenagers can connect goodwill around the world as well as any skills eg medical or green energy the island most urgently need. Generations ago, Singapore did something different; its 6 million person poluation saw itself as at the cross-seas of world's first superport. It also gave back to region asean encouraging celebration of every peoples cultures and arts. It has aimed to be the 21st C most intelligent isle- where education is transformed by every 2nd grade teacher being as curious about what will ai do over the next 5 years as anyone else. Taiwan, addmitedly a 20 million person island, chose 1987 to become world number 1 as chip design changed to maximise customer requirements instead of the moores law era where at most one new chip a year would be designed in line with Intel's 3 decades of promising 100 times more capacity every decade.

In 2025, the vibrant aAInations index is one way of looking at where is place being led to maximise its peoples intelligence opportunities for evryone to win-win (network entreprenurially)

Happy 2025- free offer first quarter of 2025 - ask us any positive question about von neumann's purpose of intelligence/brainworking - by April we hope there will be a smart agent of neumann! - chris.macrae@yahoo.co.uk

Maths-Lab-Crisis.docx

Joun in perplexity chats 

Does AI have name for terrifying ignorance rsks eg Los Angeles failed insurance sharing

In these days of LLM modeling, is there one integral one for multilateral systems reponsibilities

Is Ethiopia's new secirity model an Africawide benchmark

can you hlep map womens deepest  intel nets

what can you tell us about ...


thanks to JvN

2025report.com aims to celebrate first 75 years that followers of Adam Smith , Commonwealth begun by Queen Victoria, James Wilson and dozens of Royal Societies, Keynes saw from being briefed 1951 by NET (Neumann Einstein Turing). Please contacts us if you have a positive contribution - we will log these at www.economistdiary.com/1976 www.economistdiary.com/2001 and www.economistdiary.com/2023 (admittedly a preview!!)

First a summary of what the NET asked to be meidiated to integrate trust during what they foresaw as a chaotic period.

Roughly they foresaw population growth quadrupling from 2 billion to 8 billion

They were most concerned that some people would access million times moore tech by 1995 another million times moore by 2015 another million times moore by 2025. Would those with such access unite good for all. If we go back to 1760s first decade that scots invented engines around Glash=gow University James Wat and diarist Adam Smith we can note this happened just over a quarter of millennium into age of empire. WE welcome corrections be this age appears to have been a hectic race between Portugal, Spain, France Britain Netherlands as probbly the first 5 to set the system pattern. I still dont understand was it ineviatble when say the Porttuguese king bet his nations shirt on navigation that this would involve agressive trades with guns forcing the terms of trade and colonisation often being a 2nd step and then a 3rd steb being taking slaves to do the work of building on a newly conquered land. I put this way because the NET were clear almost every place in 1951 needed to complete both independence and then interdependence of above zero sum trading games. Whils traidning things runs into zero sums (eg when there is overall scarcity) life critical knowhow or apps can multiplu=y value in use. Thats was a defining value in meidting how the neyt's new engineering was mapped. Of course this problem was from 1945 occuring in a world where war had typiclly done of the following to your place:

your capital cities had been flattened by bombing - necessitating architecture rebuild as well as perhaps an all chnage in land ownership

your peoples had gone through up to 6 years of barbaric occupation -how would this be mediated (public served) particularly if you were a nation moving from radio to television

yiu mifgt eb britain have been on winning side but if huge debt to arms you had bought

primarily you might be usa now expected by most outside USSR to lead every advance'

in population terms you might be inland rural (more than half of humans) where you had much the least knowledge on what had hapened because you had been left out of the era of connecting electricity and communications grids

The NETts overall summary : beware experts in energy will be the most hated but wanted by national leaders; and then far greater will be exponential risk is the most brilliant of connectors of our new engines will become even more hated and wanted. We should remember that the NET did not begin with lets design computers. They began with Einstein's 1905 publications; newtonian science is at the deepest limits systemically wrong for living with nature's rules.

WE can thrash through more understanding of how the NET mapped the challenges from 1951 at https://neumann.ning.com/ Unfortunatnely nobody knew that within 6 years of going massively public in 1951 with their new engineering visions, all of the net would be dead. One of the most amzaing documents I have ever seen is the last month's diary of von neumann roughly October 1955 before he became bedridden with cancer. All over usa engineering projects were receiving his last genius inputs. And yet more amazing for those interested in intelligence machines is his last curriculum the computer and the brain scribbled from his bedroom in bethesda and presented posthumously by his 2nd wife Klara at Yale 1957 before she took her own life about a year later. A great loss because while neumann had architected computers she had arguably been the chief coder. Just to be clear Turing also left behind a chief coder Jane who continued to work for Britain's defence planning at cheltenham for a couple of decades. Economistwomen.com  I like to believe that the founders of brainworking machines foresaw not only that women coders would be as produytive as men but that they would linking sustainability from bottom up of every community. At least that is a valid way of looking at how primarily 1billion asian women batted the systemic poverty of being disconnected from the outside world even as coastal places leapt ahead with in some cases (G Silicon Valley, whatever you call Japan-Korea south-Taiwan-HK-Singapore access to all of 10**18 times moore

Epoch changing Guides

1 AI Training AI Training.docx

 2 Exploring cultural weaknesss of encounters with greatest brain tool.docx

.2016-23.pptx

help assemble 100000 millennials summitfuture.com and GAMES of  worldrecordjobs.com card pack 1 i lets leap froward from cop26 glasgow nov 2021 - 260th year of machines and humans started up by smith and watt- chris.macrae@yahoo.co.uk-

WE APPROACH 65th year of  Neumann's tech legacy - 100 times more tech decade - which some people call Industrial Rev 4 or Arttificial Intel blending with humans; co-author 2025report.com, networker foundation of The Economist's Norman Macrae -

my father The Economist's norman macrae was privileged to meet von neumann- his legacy of 100 times more tech per decade informed much of dad's dialogues with world leaders at The Economist - in active retirement dad's first project to be von neumanns official biographer - english edition ; recently published japanese edition - queries welcomed; in 1984 i co-authored 2025report.com - this was celebrating 12 th year that dad( from 1972, also year silicon valley was born) argued for entrepreneurial revolution (ie humanity to be sustainable would need to value on sme networks not big corporate nor big gov); final edition of 2025report is being updated - 1984's timelines foresaw need to prep for fall of brlin wall within a few months; purspoes of the 5 primary sdg markets were seen to be pivotal as they blended real and digital - ie efinance e-agri e-health e-learning and 100%lives matter community; the report charged public broadcasters starting with BBC with most vital challenge- by year 2000 ensure billions of people were debating man's biggest risk as discrepancy in incomes and expectations of rich & poor nations; mediated at the right time everyone could linkin ideas as first main use of digital webs--- the failure to do this has led to fake media, failures to encourage younger half of the world to maxinise borderless friendships and sdg collabs - see eg economistwomen.com abedmooc.com teachforsdgs.com ecop26.com as 2020s becomes last chance for youth to be teh sustainability generation


 

© 2026   Created by chris macrae.   Powered by

Report an Issue  |  Terms of Service