Thread Guy

Thread Guy × Nick Bruhman

Sneaker Twitter became a live laboratory for automated market capture, where 1–2-second advantages turned bots, proxies, and software licenses into speculative assets worth up to $30,000.The ecosystem monetized every layer of scarcity, but collapsed when post-COVID liquidity and demand disappeared; its unusually fast operators then migrated into crypto, startups, streaming, and infrastructure, including Slash.

Thread Guy

Flood says Situational Awareness correctly identified AGI-driven demand for compute, energy, and infrastructure, but positions nearing 20% of some companies’ market caps made the trade difficult to exit.Leverage, falling prices, and rising volatility triggered billions in same-day collateral demands, reportedly forcing a distressed sale to Citadel, while Jane Street’s absence remains unexplained.With roughly $1 trillion of hyperscaler spending still lacking clear returns, Flood holds about 50% cash as he waits to see whether AI can show the money.

Thread Guy × Bubble Boi

Kimi’s linear-attention architecture may keep most model memory fixed as context grows, weakening Bubble Boi’s memory-and-capex thesis for AI winners.That could reduce future DRAM and rack demand while improving premium Nvidia GPU economics; he favors selectively priced CXMT, remains bullish on Intel’s 14A customer commitment, and is watching neoclouds, flash-based inference, and Qualcomm’s unproven throughput claims.

Thread Guy × GoodAlexander

SpaceX’s roughly $25 billion IPO, with about 30% allocated to retail, could drain liquidity from adjacent speculative assets even if the company trades well.GoodAlexander prefers OpenAI because 900 million ChatGPT DAUs and a reportedly $100 million six-week ad pilot create an Amazon-like monetization option, while orbital-compute execution, legacy-telecom disruption, and user trust remain key risks.

Thread Guy × Andrew Kang

Andrew Kang says robotics is shifting from mechanical design toward intelligence, with reliability now the next threshold for physical intelligence.He expects GPT-5-level robots in roughly a year and home-capable robots in about two years, while hardware, manufacturing, supply chains, model commoditization, and labor displacement shape the opportunity.

Thread Guy × Bubble Boi

Intel’s turnaround thesis is shifting from fabs alone to advanced packaging, with improved 18A yields and a possible 14A leapfrog over TSMC if execution continues.EMIB could enable denser, more repairable multi-die systems, while token demand and semiconductor capex support at least three more years of AI infrastructure spending; execution and speculative bottleneck trades remain the risks.

GoodAlexander × David Choi

Chinese open-source models reportedly approach Opus or Sonnet performance on roughly $10,000 GPUs, while Qwen 3.6-class compute costs about $2 per hour.That 40-80x gap supports local inference and crypto-financed compute, but USD.AI’s $1 billion GPU-loan target, reference rate, and CHIP model remain unproven.

Cobie × Thread Guy

Stablecoins, Polymarket, Hyperliquid and TradeXYZ show crypto rails gaining adoption despite waning attention to many investable assets.Tokenized equities, commodities and perpetuals could attract users and capital, with TradeXYZ reportedly forecasting Monday’s open within roughly 50 basis points.A new speculative cycle may follow, but AI-driven wealth concentration and private-market inequality remain unresolved risks.

Andrew Yang × Thread Guy

Yang sees economic mobility narrowing from roughly 93% for Americans born in the 1960s to 50% for those born in the 1990s.His “lead and be led” playbook turns time, attention and social capital into economic leverage, while AI may eliminate many more jobs than it creates.The catalyst is whether taxation and shared proceeds can make AI visibly beneficial before “AI plus capitalism” produces a “loot and rob” equilibrium.

Thread Guy × Mark Jeffrey

Bittensor is framed as Bitcoin’s programmable mining layer, with TAO subsidizing AI and digital-commodity subnets.Templar subnet 3 claims decentralized training of a 72-billion-parameter frontier model; Ridges reportedly built a Claude-and-Cursor competitor for $1 million-$2 million at $29 monthly, while regulation, hacks, and protocol design remain risks.