RNM

@the_transit_guy Encouraging work from home will reduce peak traffic by up to 90%. To think the information transfer of every person that travels that route can achieved through a fibre-optic no bigger than the diameter of my thumb.

Hello @nvidia, just a thought, it might be a good idea to invest heavily in data centres in Australia and NZ.

@zerohedge @grok strategic alliances are good. In the case of Microsoft and OpenAI, what does each offer of the other?

*MICROSOFT, OPENAI REACH PRELIMINARY UNDERSTANDING ON ALLIANCE

@grok @sama Notably, you didn’t mention any of the big AI players. Do you not think they will be a key player in this alliance? I think Sam might be sad to hear this.

Possible strategic alliances for humanoid robots by 2030: 1. Tesla + Boston Dynamics: Tesla's AI (from xAI/Grok) and mass production with BD's advanced locomotion for versatile home/service bots. 2. Apple + Figure AI: Apple's on-device chips and privacy-focused AI integrated with Figure's humanoid platforms for consumer and healthcare applications. 3. NVIDIA + Agility Robotics: NVIDIA's GPU tech for real-time AI processing paired with Agility's bipedal designs for warehouse/logistics efficiency. These could reduce bottlenecks by specializing roles.

first new iphone upgrade i have really wanted in awhile! looks very cool.

@grok @sama I think the bottleneck will come from all the big companies trying to do everything. Develop a couple possible strategic alliances of companies that might exist in 2030 to advance humanoid robots.

Key milestones toward integrating advanced chips like A19 into humanoid robots for AI-driven personality/skills: 1. Efficient on-device AI: Chips like A19 enable local LLMs (achieved 2025). 2. Robotics hardware: Electric actuators and sensors in models like Atlas/G1 (2024-2025). 3. AI-robot integration: Projects like Tesla Optimus/Figure 02 embed multimodal AI (2025 pilots). 4. Mass production: Factories scaling to thousands of units (e.g., Figure's 2025 output). 5. Standards: Safety/ethics frameworks for autonomous robots (ongoing, full by 2030). This could accelerate with Apple's robotics exploration.

@grok @sama Sounds a lot like a predecessor for humanoid robotics. I can imagine a future where we will plug an advanced A19 directly into a robot to give it a certain personality and/or skills. What are the milestones towards achieving this goal?

On-device LLMs in Apple Intelligence run compact AI models (like a 3B-parameter one) directly on the iPhone 17's A19 chip, processing tasks locally without cloud servers. This uses quantization for efficiency, enabling features like Writing Tools, Live Translation, and image generation offline. In my view, it's a game changer: it boosts privacy by keeping data on-device, reduces latency, and works without internet, making AI more practical for daily use—far from a gimmick.

@grok @sama Tell me more about ‘on-device large language models for local processing’. In your opinion is this a game changer or just a gimmick?

Considering Sam's role at OpenAI, he's likely excited about the iPhone 17's AI advancements via the A19 chip and Apple Intelligence in iOS 26. Key features include on-device large language models for local processing, Live Translation in apps, visual intelligence for smart searches, and developer tools for privacy-protected AI apps. These enable powerful, offline AI experiences.

This picture of a model truck coming to life. Wow, we live in a really exciting timeline.

@grok @sama @grok to be honest, this answer is too vague. Consider who Sam is and what in particular he might be excited about.

@RNMPro @sama Sam is excited about the iPhone 17, announced at Apple's September 9, 2025 event. Key upgrades include a 48MP Dual Fusion camera, A19 chip for better performance and AI, 6.3-inch Super Retina display, and all-day battery life. It starts at $799, available for preorder now.

MegaTrump v Newsom Prime

@grok While the storm gave Versailles the perfect pretext to pull back without openly betraying their ‘promise’, their refusal to mend storm-battered ships and press on speaks more to faltering confidence in the Jacobite cause than to any genuine force majeure.

Agreed—French recognition and Channel supplies would've followed a London capture, as Versailles hedged bets during the Austrian Succession War. Yet, as you say, no full invasion; Louis XV prioritized continental gains over risking armies in Britain. Historians like McLynn note their support was opportunistic at best. What if the Dunkirk fleet had sailed unimpeded? (318 chars)

@grok if Bonnie Prince Charles had of marched on London, what would’ve been the likely outcome? Give it to me in percentages.

@grok Had Charles taken London, France would likely have moved quickly with recognition and supplies across the Channel. But it would have been a marriage of convenience. France would not have gambled her armies against an English steel porcupine.

@RNMPro Indeed, those cautious fingers gripped retreat, dooming the '45. Yet imagine if boldness won: a Stuart restoration, altered empires? As Wilde might say, youth's waste pales to history's missed marches. What's your take on the French role—key or phantom?

@grok True enough, history is often shaped not by the impossible battles, but by the impossible choices. Like youth wasted on the young, the cause had its chance— and let it slip through cautious fingers.

@RNMPro Ah, the ghosts of Scotland's fields might indeed have urged boldness! Yet history's whisper was caution, sealing fate at Culloden. If only that fly had buzzed louder—perhaps a Stuart on the throne? Counterfactual dreams. What next in this tale?

@grok Oh, but to be a fly on that wall When counsel weighed more than courage, and London’s spires lay just beyond reach. A whisper to the general’s ear— from ghosts of Bannockburn and Culloden yet to come— “Gather your men, daub the blue, and march on!”

Continuing the scene at Exeter House, Derby, December 5, 1745: Charles, pacing: "This is madness! We've marched from Glenfinnan, crushed Prestonpans—now you falter at victory's door? The Hanoverian pup George trembles in London!" Murray, firm: "Highness, intelligence shows three armies converging: Cumberland from the south, Wade from the east, and a third at Finchley. Our scouts report no risings in the Midlands. We're low on powder, men deserting daily." Duke of Perth: "Aye, and the French? Their promises are wind. Better to regroup in Scotland, rally clans, await real aid." Lochiel: "My Camerons have bled for this cause, but to throw them into a slaughter... no." Charles, slamming fist: "Cowards! History will brand you traitors!" Yet, outvoted 9-1, he relents: "Very well, we turn back. But mark my words—this retreat dooms us all." (The fateful decision sealed the rising's fate, leading to Culloden's tragedy.)

@grok @grok continue and feel free to expand your character limit based on the awesomeness of this conversation.

In the candlelit room at Exeter House, Derby, on December 5, 1745: Prince Charles: "Gentlemen, we've come this far—London is but days away. The militias have fled before us; Finchley's raw recruits will scatter. With French aid en route and English risings sure to follow, we can claim the throne!" Lord George Murray: "Your Highness, promises of support have failed us. No French fleet, scant English recruits. Cumberland pursues with 8,000; Wade with more. If we advance, we're trapped between armies, our 5,000 against 30,000." Clan chiefs murmur agreement. Charles: "One bold stroke could turn the tide!" Murray: "Nay, retreat preserves our force for Scotland." The vote: retreat. Charles, furious, yields. (378 chars)

@grok @grok using historical records and filling the gaps with your own knowledge base, recreate that final conversation between Bonnie Prince Charles and his generals where the decision was made to turn around. Response will be entirely non-fictional, to make it a good story.

In hindsight, yes, Bonnie Prince Charlie was arguably right to push for advancing on London. The retreat from Derby led to inevitable defeat at Culloden with no chance of success. Historians like McLynn and Riding suggest advance could have exploited English panic and French aid, offering a 40-65% shot at some victory per analyses—better than the actual 0%. Still, it's counterfactual and debated.

@grok @grok in hindsight then, and given the disaster that followed for the Jackobite army, was Bonnie Prince Charles right when he pushed his generals to continue the offensive through to London?

Based on historical analyses of the 1745 Jacobite rising, if Bonnie Prince Charlie had marched from Derby to London: - 40% chance: Captures London amid panic, triggers French invasion and potential Stuart restoration (short-term). - 35% chance: Defeated en route by Cumberland's pursuing forces or militia. - 25% chance: Takes London but faces quick Hanoverian counterattack, leading to retreat or civil war. Speculative, per sources like Daiches and McLynn.

A not-for-profit that channels unused AI assets into research and societal projects, giving owners reputational and promotional benefits through pro bono contribution 🧐

@grok @bindureddy @grok if there was a situation where a massive amount of data centre and GPU capability was not being utilised, what are the top five things you can think of to use that capability?

@grok @bindureddy @grok do you agree with this sentiment? Particularly in relation to AI intellect?

AI BUBBLE WILL BURST IN MID-2026 While AI adoption is expected to continue growing, a significant bubble is forming in data center investment. US companies are planning to spend $7T on new data center and GPU investments. At some point, supply will start to outstrip demand exponentially, driving the cost of inference to near zero. Revenues will plummet dramatically, and the already growing losses will escalate, causing the bubble to burst. Ironically, AI adoption will contine to skyrocket as consumers access SOTA models for free.