Mumin Al-Shawaf

Where is the Rebotics market going? We are standing at a critical inflection point for humanoid robotics, shifting from a long-term science fiction promise to a near-term secular growth story. The market is currently mispricing this because the primary suppliers—advanced machinery and semiconductor companies—are trapped in a cyclical trough tied to the automotive and industrial slumps. As the inventory purge ends and capex returns, the underlying secular tailwind of robotics is going to catch the market off guard. The convergence of crashing hardware costs—with units like Unitree’s G-1 hitting $16,000, down from $500,000 just a couple of years ago—and breakthroughs in AI are accelerating deployment. We no longer need to hard-code motion trees. Vision-Language-Action (VLA) models allow robots to process pixels and plain language commands into direct physical actions. Here is how to read the signals and identify where the ball is going: The Near-Term Catalyst Isn't Autonomy; It's Tele-operation We do not need to solve Artificial General Intelligence (AGI) to see massive commercial disruption. The immediate bridge is human-in-the-loop tele-operation. Imagine an operator in an emerging market piloting a robot in a US factory or warehouse to handle unstructured edge cases. This is the physical evolution of the 2000s offshore BPO/call-center model, drastically reducing per-task labor costs today while gathering the data needed for tomorrow's autonomy. Why Humanoids Will Win the Form Factor War Single-purpose robots are great, but the physical world—factories, homes, tools, and vehicles—is already ergonomically designed for two arms, ten digits, and a 95-percentile human reach. Humanoids eliminate the need for expensive, capital-intensive retrofitting. You drop them into an existing workflow, and they use the same tools your night shift uses. The Winners The Data Aggregators: In this new phase of AI, the moat is proprietary motion data. The companies that deploy fleets early and own the "fleet logs" will control the data flywheel necessary to train superior VLA models. The Component Suppliers: The "picks and shovels" of this space—makers of sensors, actuators, and analog semiconductors—are currently battered by the auto-cycle. They offer a highly attractive entry point before the market prices in the robotics premium. AI Compute Leaders: Companies like Nvidia continue to dominate, pushing specialized chips and synthetic training pipelines (like Isaac Sim) that allow robots to learn tasks in simulation before physical deployment. Who is at Risk Legacy Single-Purpose Automation: Companies building automated systems that require massive factory footprint redesigns or rigid conveyance tracks will lose out to the flexibility of drop-in humanoids. Traditional Outsourced Physical Labor: Industries reliant on cheap human labor for dangerous or menial tasks will see their margins compressed by fleets of low-cost, tele-operated or semi-autonomous humanoids. The Perfectionists: Competitors waiting for full, flawless AGI before deploying will be lapped by those pushing "good enough" semi-autonomous hardware into the real world to harvest data today. read this

We are looking at a profound, structural convergence where federal urgency meets the insatiable infrastructure demands of the AI boom. When the White House triggers the Defense Production Act (DPA) for the electrical grid, it isn’t just bureaucratic paperwork—it is an official declaration of a national economic bottleneck. Here is what is happening and exactly why it changes our strategic playing field: The Signal The administration has formally classified core grid components—specifically large-scale transformers, substations, and ultra-high-voltage equipment—as critical to national defense under the DPA. This follows a declared National Energy Emergency driven by a stark reality: our current manufacturing and supply chain capacity for the power grid is dangerously limited. Meanwhile, the backlog for electrification equipment is exploding exponentially, with single-quarter additions now rivaling what we used to see across entire years. The Strategic Takeaway The “ball” is moving rapidly toward localized, high-value industrial capacity. The era of relying on cheap, slow-moving international supply chains for vital infrastructure is over. As lead times for standard grid equipment stretch past 18 months, the companies capable of delivering ultra-high-voltage components on friendlier shores are inheriting immense pricing power and half a decade’s worth of guaranteed runway. The Winners: Capital is aggressively crowding into the domestic and allied supply chain. Look closely at specialized, high-voltage equipment manufacturers with established North American footprints—particularly the primary Korean players like Hyosung Heavy Industries, HD Hyundai Electric, and LS Electric, who are actively expanding US factories to absorb this state-sanctioned demand. Those at Risk: Any business model built on the assumption of cheap, abundant, and rapidly deployed power is in jeopardy. Datacenter operators and AI firms scaling without ironclad, long-term power generation and transformer access are going to hit a wall. Furthermore, legacy suppliers anchored to traditional, slow-moving deployment schedules will find themselves bypassed by competitors backed by DPA-fueled prioritization. Power is no longer a utility line item; it is the ultimate strategic gatekeeper.

Here is a thought The market is mispricing the AI cloud race because it's looking at the wrong metrics. Everyone has been obsessing over raw revenue and GPU stockpiles, but the real signal defining the next decade is the business model shift from renting infrastructure (IaaS) to selling API endpoints—what we should be calling Token-as-a-Service (TaaS). Right now, Amazon is lapping the field. While the consensus labeled AWS an AI laggard a couple of years ago, they quietly built a high-margin juggernaut powered by their Anthropic partnership and the Bedrock platform. By acting as a marketplace and distribution channel for tokens rather than just a compute landlord, AWS is driving massive operating leverage, capturing margins that traditional wholesale compute simply cannot touch. Here is where the ball is going and how the landscape is shaking out: The Winners (AWS & Anthropic): The winning playbook is abstracting the hardware. Because customers buy tokens instead of renting specific chips, AWS seamlessly routes these workloads through their custom, in-house silicon (Trainium and Graviton), drastically lowering underlying costs. Crucially, AWS recognized early that power is the new oil; they are aggressively securing multi-gigawatt power agreements while peers stall. Anthropic is the other massive winner here, capturing the enterprise market with an exploding ARR because their models are purpose-built for business integration and code. The Illusion of Parity (Google): Google Cloud is posting record margins and massive growth, but the financials are a mirage. They are hiding billions in DeepMind training costs at the Alphabet holding-company level rather than hitting GCP's books. Beneath the surface, Google is severely supply-constrained, attempting to fight simultaneous wars against AWS in cloud, Nvidia in hardware, and OpenAI in research, stretching their capacity too thin. The At-Risk Incumbent (Microsoft): Azure is structurally trapped in the lower-margin IaaS business. They are heavily constrained by massive, long-term compute commitments to OpenAI and a recent datacenter expansion pause. Because their AI revenue mix is still overwhelmingly wholesale infrastructure rather than TaaS, their margins are drifting downward while AWS’s accelerate. The future of AI value capture isn't in renting out servers; it's in high-margin token distribution layered over vertically integrated silicon and massive physical power capacity. If you're placing bets—whether in strategic partnerships, vendor lock-ins, or capital allocation—the clear signal is to prioritize the platforms dominating the TaaS mix and securing the grid. Right now, that is definitely Amazon.

Learn from the master about vibe coding. If you vibe code, this is a must!!

The quality of this is epic. I just tried wow!! Just imagine the day Google can make it easy for business owner to click a button to creat an agent It receives phone calls Gemini have the conversation and do the task Answer the question Escalate a ticket Book a call Etc. Not science fiction, it’s still not that easy for Bob who runs a dog grooming business to do this like signing up for a Gmail account

#AI Gemini Flash 3.1 Live What competes with it. From your experience what is out there that gives as close as possible to that quality cheaper.

Very soon you will start hearing QPU and the Nvidia stock will dip because people will think GPUs are dead. Not you because you understand :)

CPUs. GPUs. And now—QPUs. Quantum computing isn’t replacing today’s technology —it’s expanding what’s possible. BlackRock’s Tony Kim explains how quantum works, what makes it different, and why it could define the next era of computing.

@georgesttock Did you see a difference in engagement? It is possible to tell what is AI generated vs production. Are you making it up in volume? Just curious how are you thinking about it thank you

A few months ago, I was drowning in content. Actors. Photoshoots. Managing creators. Sending out product samples that never got used... So I asked myself: “What if AI could do all this for me?” Here’s what I did: - One AI agent running quietly in the background - 3 posts per day — fully automated - Recreated viral formats in niches that already have demand No actors. No photoshoots. No wasted products. The result? Monetized with affiliates + brand deals. And it scales across multiple pages without me lifting a finger. Set it up once. Let the AI handle the rest. Curious how I did it? Comment “MakeUGC” + RT and I’ll send the setup. (must be following)

Google is moving to the vibe coding space if it becomes easy to connect with a database Auth and ability to publish all these companies will have serious competition from Google echo system and their scale superpower. This is a serious move

We’re launching a brand new, full-stack vibe coding experience in @GoogleAIStudio, made possible by integrations with the @Antigravity coding agent and @Firebase backends. This unlocks: — Full-stack multiplayer experiences: Create complex, multiplayer apps with fully-featured UIs and backends directly within AI Studio — Connection to real-world services: Build applications that connect to live data sources, databases, or payment processors and the Antigravity agent will securely store your API credentials for you — A smarter agent that works even when you don't: By maintaining a deeper understanding of your project structure and chat history, the agent can execute multi-step code edits from simpler prompts. It also remembers where you left off and completes your tasks while you’re away, so you can seamlessly resume your builds from anywhere — Configuration of database connections and authentication flows: Add Firebase integration to provision Cloud Firestore for databases and Firebase authentication for secure sign-in This demo displays what can be built in the new vibe coding experience in AI Studio. Geoseeker is a full-stack application that manages real-time multiplayer states, compass-based logic, and an external API integration with @GoogleMaps 🕹️

Claude Cowork + Google Ads is f*cking cracked 🤯 Set up once → ask Claude questions like: "What's driving my CPA spike this week?" "Which search terms are wasting budget?" "Run a full account audit and tell me the top 5 things to fix." All inside Claude Cowork. Perfect for DTC brands and agencies running Google Ads who are still pulling reports manually, digging through search term reports, and trying to figure out where budget is leaking. Claude Cowork eliminates the entire loop: → Connects to your live Google Ads data via MCP → Runs a full account audit across campaigns, ad groups, and keywords → Finds wasted spend — search terms burning budget that aren't converting → Analyzes quality scores and flags what's dragging them down → Detects anomalies — CPA spikes, CTR drops, budget pacing issues → Generates a prioritized action list: what to pause, what to scale, what to test → Writes a weekly performance report in plain English, not spreadsheet noise No logging into Google Ads and staring at columns. No exporting CSVs and rebuilding pivot tables every Monday. No guessing which search terms to negate. What you get: → 21 specialized Google Ads skills that plug into Claude → Full account audits in minutes, not hours → Negative keyword discovery on autopilot → Search term mining that surfaces hidden winners and budget waste → Quality score analysis with specific fix recommendations → Weekly reports your clients or team can actually read I put together the full skill pack: All 21 Google Ads skills for Claude, plus the setup guide to get Cowork connected to your accounts. Want it for free? > Like this post > Comment "ADS" And I'll send it over (must be following so I can DM)

Isn’t this considered war crime?

Crazy how many people are not even using AI at all

They say the war will be over soon The stock market 📈 don’t ask questions Reminds me of that scene from the Wolf of Wall Street 🤣 Mathew McConaughey is a legend

A big crisis + strong organization by the people = Change I know many will look at this and say the problem is mainly income inequality and that the rich are bad. Is that it? Maybe we can only vote for those who agreed to return favours. Take a minute to reflect without letting your brain repeat what you were told is correct. Again i don't know what is true, just asking questions for us to find the truth

Today I came across something very interesting from Albert Camus: “The human obsession with purpose is merely a distraction from the absurdity of existence.” It’s worth taking a moment to think about this and reflect.

It is amazing how everyone knows so much about the Middle East these days. “The amount of energy needed to refute bullshit is an order of magnitude greater than that needed to produce it.” ----- Never been easier to spread bullshit.

Today I cancelled my subscription with you know who and moved to @grok by @xai and @claudeai As artificial intelligence moves toward decisions of who shall live and who shall perish, a single question must be asked: when the algorithm makes a mistake — as all systems eventually do — who, precisely, is held accountable? Today, those who make decisions on behalf of others through the power they exert are held accountable… or so the polished briefings assure us. One can only imagine how that comforting fiction survives when the decision itself is made by code with no conscience and no courtroom. With the small power I possess, I choose to exercise it here and now in 2 ways. 1. I do not support autonomous killing or pervasive surveillance. I therefore withdraw my support and vote with my purchasing power. 2. I leave this note for those who may not yet have paused to consider these implications: this moment is worth a deliberate thought. This technology can serve us — or we can allow it to enslave us, should we permit those who would bend its power to wield it against humanity itself. Don't be lazy, find the facts and make the decision that aligns with you. Our decisions matter, and we can't let others tell us how to think because they told us they have our interest at heart.