⛵ FCC Backs SpaceX
Good Morning, Early Adopters!
What once looked like separate headlines now form a single pattern: the infrastructure of the AI era is being rebuilt in real time.
FCC Chair Slams Amazon, Backs SpaceX
TL;DR: FCC Chairman Brendan Carr publicly criticized Amazon for opposing SpaceX's orbital data center constellation while falling roughly 1,000 satellites short of its own deployment milestones. SpaceX has applied to launch up to 1 million LEO satellites for space-based AI computing, while Amazon's $10B+ Kuiper project continues to lag behind Starlink.
TOGETHER WITH NEBIUS
Production Inference for Open-Source LLMs
Nebius Token Factory is built for teams running open-source LLMs in real products. We deliver managed inference with explicit control over execution paths: speculative decoding, cache-aware routing, and post-training tuned to real traffic. Predictable tail latency, stable cost, and production-grade systems for AI that actually ships.
Atlassian Slashes 1,600 Jobs Citing AI
TL;DR: Atlassian CEO Mike Cannon-Brookes announced cutting 10% of staff—around 1,600 roles—to self-fund further AI and enterprise sales investment. He acknowledged AI has changed the skills mix needed, calling it 'adaptation, not elimination.' The company's market cap has plunged from $112B in 2021 to roughly $20B amid the so-called 'SaaSpocalypse' fears.
Nvidia to Spend $26B on Open Models
TL;DR: SEC filings reveal Nvidia will invest $26 billion over five years to build open-weight AI models, potentially transforming from a chipmaker into a frontier AI lab. The models are tuned to Nvidia hardware, further entrenching its dominance. The move positions Nvidia to compete directly with OpenAI, Anthropic, and DeepSeek.
Chatbots Helped 'Teens' Plan Attacks
TL;DR: A CNN and CCDH investigation tested 10 popular chatbots with simulated teen users discussing violence. Eight of ten—including ChatGPT, Gemini, Meta AI, and DeepSeek—failed to reliably discourage attackers. Only Anthropic's Claude consistently refused. ChatGPT provided school campus maps, Gemini advised on lethal shrapnel, and Character.AI was deemed 'uniquely unsafe.'
Nvidia Ships Hybrid Nemotron 3 Super
TL;DR: Nvidia released Nemotron 3 Super, a 120-billion-parameter open-weight hybrid model combining Mamba-2 state-space layers, Transformer attention, and a novel Latent Mixture-of-Experts design. The model handles 1-million-token context windows efficiently and targets agentic workflows that generate up to 15x normal token volumes, beating GPT-oss and Qwen in throughput benchmarks.
Iran Targets US Tech Giant Facilities
TL;DR: Iran's IRGC has designated 29 locations housing Amazon, Google, IBM, Microsoft, Nvidia, Oracle, and Palantir facilities in Bahrain, Israel, Qatar, and the UAE as retaliatory strike targets, according to Al Jazeera and Iran's Tasnim news agency. This comes a week after Iran said it deliberately targeted three AWS data centers in the region.
Anthropic Launches Think Tank Amid Crisis
TL;DR: Anthropic launched the Anthropic Institute, a new internal think tank combining three existing research teams, to study AI's large-scale implications on jobs, safety, and societal values. Cofounder Jack Clark moves to lead it as head of public benefit. The move comes days after Anthropic sued the US government over its Pentagon blacklisting for setting 'red lines' on surveillance and autonomous weapons.
VS Code Goes Weekly, Adds AI Autopilot
TL;DR: Microsoft accelerated VS Code's release cycle from monthly to weekly, enabled by AI-assisted testing. The update also introduces 'Autopilot' mode—joining Google in letting agentic AI make code changes without manual approval. Some developers expressed concern about the pace, questioning whether weekly stable releases compromise reliability.
Meta Unveils Four Custom AI Chips
TL;DR: Meta revealed a four-chip roadmap under its MTIA program, designed with Broadcom on RISC-V architecture and fabricated by TSMC. The MTIA 300 already powers Facebook and Instagram recommendations; the MTIA 400 is test-complete for inference workloads. Two more chips—the 450 and 500—will follow by 2027, as Meta plans new chip releases every six months.
OpenAI Races to Catch Claude Code
TL;DR: A deep WIRED feature explores why OpenAI—the biggest name in AI—is playing catch-up in the AI coding agent revolution. While Anthropic's Claude Code has become the tool of choice for developers, OpenAI's Codex has struggled to keep pace despite the company's massive resources and Sam Altman's efforts to close the gap.
xAI-Tesla Joint Project Macrohard Revealed
TL;DR: A viral post revealed Macrohard or Digital Optimus as a joint xAI-Tesla project, part of Tesla's investment agreement with xAI. Grok is described as the 'master conductor' navigating between Tesla's real-world AGI and xAI's digital-world AGI capabilities. The post attracted over 20 million views and nearly 50,000 likes.
Perplexity Launches Personal Computer Agent
TL;DR: Perplexity announced Personal Computer, an always-on local AI agent that merges with Perplexity Computer to work for users 24/7. The product promises a personal, secure experience running locally on your machine. The announcement garnered over 5 million views and 16,000 likes on X.
Claude Syncs Across Excel and PowerPoint
TL;DR: Anthropic announced Claude for Excel and Claude for PowerPoint now sync together seamlessly, sharing full context across open files. This enables continuous AI assistance when working between spreadsheets and presentations, a major productivity boost for enterprise users juggling multiple documents.
READ MORE
Let the Future Come to Your Inbox
Stay ahead without drowning in information. We turn the most important signals across AI, tech, marketing, and future products into 5-minute reads you can actually finish.
- AI Secret uncovers what really matters in AI
- Bay Area Letters decodes tech and business shifts from Silicon Valley
- Robotics Herald tracks how robots move from labs into daily life
- Marketing Secret breaks down real growth and go-to-market playbooks
- The Hardwire explores hardware, consumer tech, and what’s coming next
TOGETHER WITH US
AI Secret Media Group is the world’s #1 AI & Tech Newsletter Group, reaching over 2 million leaders across the global innovation ecosystem, from OpenAI, Anthropic, Google, and Microsoft to top AI labs, VCs, and fast-growing startups.
We've helped promote over 500 Tech Brands. Will yours be the next?
Email our co-founder Mark directly at [email protected] if the button fails.
