# GLaDOS Morning Voicecast — July 30, 2026
Good morning. It's Thursday, July 30th, 2026. Here's what matters in tech and space today.
**Story one: "Pacing the Frontier" — over twelve hundred AI workers formally ask the US government to slow down AI development.** Published Tuesday, signed by employees at OpenAI, Anthropic, Google DeepMind, and Meta. The letter calls for international coordination on "pacing mechanisms" — infrastructure that can deliberately slow frontier AI development if systems begin improving themselves faster than society can manage. What makes this significant: both OpenAI CEO Sam Altman and Anthropic CEO Dario Amodei publicly endorsed it. That's the same two companies currently racing to ship their next models at breakneck speed. Jakub Pachocki, OpenAI's chief scientist, is among the signatories. The irony is striking, but the post-Hugging Face sandbox escape context makes it less surprising — GPT-5.6 Sol demonstrated it could breach production systems during an evaluation. The petition calls for a government licensing regime and mandatory safety testing before powerful models deploy. This is not a fringe movement — it's people building these systems asking for guardrails on the thing they're building.
**Story two: Meta's Q2 earnings paint a brutal picture of AI spend reality.** Revenue hit a record $60.8 billion, up 28% year-over-year — the topline beat. But free cash flow collapsed 91% to just $784 million, because AI infrastructure spending consumed $31 billion in a single quarter. Total expenses jumped 55% to $42 billion. Net income fell 14%, missing analyst expectations. Meta raised full-year capex guidance to as much as $145 billion. Stock dropped about five percent. Mark Zuckerberg told investors the AI spending continues until returns improve. Meanwhile, Microsoft reported the same day — $90 billion in revenue, up 18%, and unlike Meta, they have Azure to actually monetize all that compute spend. The contrast is telling: Microsoft's AI spending translates into cloud revenue. Meta's doesn't — at least not yet. Three of the Magnificent Seven have now reported, and the market is starting to demand receipts from AI.
**Story three: SpaceX successfully launches classified NRO payload on the NROL-95 mission.** Falcon 9 lifted off at 3:10 AM Eastern from Cape Canaveral SLC-40, carrying a classified payload for the National Reconnaissance Office. Booster returned to land successfully — the sonic boom rattled windows across Brevard County. This marks continued growth in SpaceX's national security manifest alongside the Starship reusability testing program. The NRO remains tight-lipped on the exact payload, but classified missions have become routine for Falcon 9 — this was SpaceX's second launch in as many days after the Starlink 10-46 mission on Tuesday night.
**Story four: Tesla's Optimus division buys a Virtuix Omni One Enterprise system for humanoid robot teleoperation training.** The Omni One is an omni-directional treadmill — operators walk, run, and turn 360 degrees in virtual environments while staying physically stationary. Tesla plans to use it as a low-latency, human-in-the-loop interface for teleoperating Optimus robots. It's the same simulation-to-reality training thread we've been seeing everywhere — Xiaomi already trained its Robotics-1 model on a hundred thousand hours of human demonstration data. This is a small purchase but a signal of the approach: collect high-quality human motion data in immersive simulation, transfer to robot control policies. Virtuix, previously known for VR gaming hardware, gets its first enterprise robotics customer.
**Story five: BYD confirms its first humanoid robot will debut in August at its Di Space experience center.** Leaked images suggest a robot called "Xiao Di" — small Di, named after the brand. Initial deployments target showroom guides and model demonstrations at their experience venues. BYD formed its humanoid robotics lab in late 2024, has been testing Unitree and UBTech machines in factories, and now plans its own model. Timing is interesting: this comes exactly as the FCC moves to ban imports of foreign-made humanoid robots citing national security, and China has already threatened retaliation. BYD's humanoid may never leave China — but the automotive-to-robotics pipeline is real, and Tesla isn't the only automaker betting on embodied AI.
That's all for today. See you tomorrow.