Sponsored by

Welcome, busybees

Elon Musk just confirmed what a secret foundry SpaceX has been quietly building in Bastrop, Texas is actually for: manufacturing a hard-to-make gas turbine part that every AI data center builder currently depends on a tiny oligopoly to supply. If it works, it hands a Musk-controlled entity a real manufacturing edge that's difficult for well-funded but non-manufacturing rivals to copy quickly. It also means a lot more gas turbines coming online fast, and turbines already in the ground are drawing federal lawsuits and peer-reviewed health research over the pollution they emit, especially in Memphis, where the NAACP has repeatedly accused Musk's xAI of running turbines without required permits.

Meanwhile, Washington tightened restrictions on foreign robots and drones this summer, but a new analysis out today makes clear China may simply route around the wall entirely. Here's what else is worth knowing, plus real roles across engineering, design, and sales you can apply to right now.

Here’s what happened in AI today:


• Musk's secret Texas foundry aims to solve AI's chip bottleneck, and it comes with a real pollution cost.

• New US barriers on foreign robots may not hold against China's manufacturing scale.

• Sony Music and Warner just sued Anthropic, alleging a "brazen campaign" of IP theft.

• Hugging Face is selling a $399 open-source robot duck, and it's a bigger deal than it sounds.

…and a whole lot more that you can read about below.

• Musk's Secret Texas Foundry Is Built to Solve AI's Chip Bottleneck, at a Real Cost

Musk confirmed this weekend that a secret foundry SpaceX has been building in Bastrop, Texas exists to manufacture a specific, hard-to-make gas turbine part, one every major AI infrastructure builder currently sources from a tiny oligopoly of suppliers. If SpaceX pulls it off, a Musk-controlled entity gains a manufacturing capability that's genuinely difficult for well-funded but non-manufacturing competitors to replicate fast. The tradeoff: it means far more gas turbines coming online quickly, and turbines already running are drawing real scrutiny. In Memphis, where SpaceXAI has powered its Colossus data centers with gas turbines since 2024, the NAACP has repeatedly accused the company of operating without the permits or pollution controls federal law requires. University of Memphis researchers found local air pollution grew "slightly worse" because of the data center, in a neighborhood already facing heavy industrial pollution.

• New US Barriers on Chinese Robots May Not Hold

Washington tightened restrictions on foreign-made advanced robotic systems in July and August, and imposed steep tariffs on imported drones and their components, both citing national security. The drone tariffs take effect in September, with additional component tariffs following in 2027. But the world's five largest humanoid robot makers by shipment volume, AgiBot, Unitree, Galbot, UBTECH, and Leju Robotics, are all Chinese, together accounting for 86% of global shipments in the first half of 2026, according to Counterpoint. Analysts expect Chinese humanoid makers to follow the same playbook Chinese EV makers used: build scale at home, expand overseas, then establish local production, with countries facing labor shortages becoming early adopters regardless of where the hardware was built.

• Sony Music and Warner just sued Anthropic

The music publishers allege a "brazen campaign of illegally torrenting, scraping, and downloading copyrighted works," naming Anthropic, CEO Dario Amodei, and co-founder Benjamin Mann directly, and seeking up to $150,000 per infringed work.

Today's Job Board

Founding Product Engineer, Full Stack | b12 Labs
San Francisco Bay Area, California (Full-time)
Help make hard molecules for pharma and biotech at an early-stage startup.

Developer Advocate (DevRel) | Signadot
Remote (Full-time)
Grow the developer community behind Signadot's testing infrastructure for AI agents.

Founding Engineer, Frontier Tuning Platform | Uplift AI
San Jose, California (Full-time)
Build the fine-tuning platform behind foundational voice models for regional languages.

Country Launcher (IR/NL/IT/ES) | Corgi Insurance
Paris, France / London, UK / Amsterdam, NL / Dublin, IE / Barcelona, ES (Full-time)
Launch Corgi's AI-native financial infrastructure into new European markets.

Founding Product Designer | Moritz
United States (Full-time)
Design the interface for an AI-native law firm handling legal work with same-day turnaround.

Senior Frontend Engineer | Authologic
Remote (Full-time)
Build the Stripe-like infrastructure for online identity verification.

Engineering Manager | OneChronos
Remote (Full-time)
Lead engineering for the AI-matched smart market used by institutional investors.

Senior Software Engineer | Confido
United States (Full-time)
Build the AI-enabled financial automation and intelligence platform used by CPG brands.

US Account Executive | Laylo
New York, NY / Los Angeles, CA / Nashville, TN / Remote (Full-time)
Sell the CRM platform powering iconic musicians, labels, and tours.

The best voice models, now across all channels

Most CX platforms do not own the voice. They orchestrate a workflow, then call a third party for speech and transcription. Every hop adds latency, cost, and another vendor to manage.

ElevenAgents is the opposite. They make the voice models the market builds on, and ElevenAgents puts full orchestration on top. Voice, transcription, text-based chat, and reasoning run in one vertically integrated pipeline, so responses come back in <400 milliseconds and sound human, not synthetic.

Plus, you keep full control. Plug in any LLM, integrate tools, webhooks, and MCP servers, and ground responses in your knowledge base. Get an agent live in minutes, then A/B test with Experiments, enforce Guardrails, and version every change.

The payoff: more human conversations, lower latency, and far less time stitching infrastructure together. You build on the models you already trust. Pricing is transparent and flat at $0.08 per minute.

Today's Money Moves 💸

Neocloud Lambda Secures $1 Billion in Debt to Buy More Chips

Lambda secured $1 billion in debt financing specifically to purchase more chips, continuing the trend of AI infrastructure providers raising debt, not just equity, to keep pace with compute demand.

Emerald AI Raises $150M to Turn Data Centers Into Grid Assets

Emerald AI closed an oversubscribed $150 million Series A at a $1.05 billion valuation, co-led by Energize Capital and DCVC, with Nvidia, Samsung Ventures, Siemens, and GE Vernova among the participants. Its software helps AI data centers flex their power draw during grid stress instead of just consuming flat out, directly relevant given today's turbine story above.

Today's Growth Recommendation

Stop using ChatGPT for generic research. Marketing Against the Grain built 22 plug-and-play ChatGPT agents for specific marketing jobs, launch in 5 minutes, free.

Get the 22 Agents Below ↓

Holiday Creator Calendars Are Filling Up. Q4 Panic Is Optional.

Creators lock in holiday calendars 90 days out, before brands finalize commission strategy and long before Black Friday.

The 90-Day Holiday Sprint guide from Levanta helps brands get creators driving holiday demand while competitors are still recruiting, from commission structure to onboarding to scaling.

Your 90-day countdown starts now.

Skill of the Day

Turn Your Judgment Into a Reusable Skill

Most AI feedback disappears the moment the chat ends. You correct something, move on, and the next session starts from zero again. Omar Saravia argues the strongest agent workflows do the opposite: a human verifies the hard outputs, then encodes that judgment into a reusable skill or verifier, so the expertise compounds instead of getting rebuilt from scratch every time.

Here's the actual workflow:

Give the agent a real task, then review the difficult output yourself.

Explain exactly why you accepted or rejected it, focusing on the decision rule behind the call, not just this one example.

Save that rule as a checklist, skill, or verifier, and reuse it on the next run.

Removing the expert from the loop was never the goal. Making the expert's judgment compound is, so your attention goes toward the next new problem instead of re-explaining the same one.

Use this right after you've corrected or approved a piece of AI output, so the reasoning behind your call doesn't disappear once the chat ends.

❝

After I review this output, turn my corrections into a reusable checklist or verifier skill for future runs. Preserve the decision criteria, not merely this example.

Invite friends & get instant rewards

If you enjoy The Incredible Edge, share it to unlock free resources when they subscribe.