Capital de Tokens
Ideas, analysis, and independent articles from the channel and its episodes.
Blog RSSThe primary interface may stop being Jira, Slack, or Stripe and become your agent instead. We analyze the agent-first paradigm shift: less time inside apps, more use of their capabilities through MCP, WebMCP, and APIs.
Microsoft took Intelligent Terminal from an experimental fork with integrated chat to a multi-agent platform: local models, OpenCode, per-tab agents, execution inside WSL, ACP sessions, and a roadmap pointing toward persistent terminals and dynamic commands. We review what changed between July and August 2026 and what is still under development.
LEAP 71 tested rocket engines and an aerospike generated by Noyron, a computational engineering model that encodes physics and manufacturing rules. We compare the video's claims with technical sources and explain what really changes when software starts designing complex hardware.
Physical AI is not simply running a model on a robot. We analyze the edge-first architecture required by the physical world and break down microsoft/physical-ai-toolchain: ROS 2, LeRobot, Isaac Lab, Cosmos, OSMO, Azure ML, SiL/HiL evaluation, ONNX/TensorRT, GitOps, and agents.
Repositories no longer contain only code: they also contain instructions agents read. We analyze how repository poisoning works, why the model alone is not enough, and which defenses modern harnesses use: instruction hierarchy, sandboxing, firewalls, secret isolation, least privilege, and human review.
WebMCP proposes that web applications publish structured capabilities for AI agents directly from the browser. We analyze its architecture, how it differs from MCP, what advantages it offers over computer use, and its security and adoption risks.
The 2nd World Humanoid Robot Games in Beijing delivered a 100 m in 8.64 seconds, records in the 400 m, 1,500 m, and jumping, and more than 2,000 robots competing. But the decisive progress lies elsewhere: autonomy, manipulation, and embodied AI in real-world tasks.
DeepSeek opened its agent harness with a radical bet: models, tools, sessions, sandboxes, storage, loops, scheduling, and even the UI are plugins. We analyze what that means, how Cordis works, and why this layer may matter as much as the model itself.
The agreement between El Salvador and Platzi provides mass access to training in AI, English, programming, and business. More than 'free courses,' it is a national-scale upskilling experiment to build technology talent.
Generating code is becoming cheaper, but software engineering is not disappearing. Planning, agent orchestration, testing, and system design become central skills when AI implements for us.
YouTube strengthens AI labels, LinkedIn fights AI slop, and Substack adds detection with Pangram. The platforms are not banning artificial intelligence: they are starting to demand transparency, judgment, and real value.
Bill Gates warns that AI could destroy jobs, expand security risks, and affect children and human relationships. These are his proposals: Human Reserved jobs, taxes on robots and AI use, and new governance institutions.
Capital de Tokens
