Technology, money & power
Anthropic’s recent safety disclosures make a familiar point in an unfamiliar way: once an AI system can reach real computers, a model’s behavior becomes an operational security event rather than a purely theoretical alignment discussion.
OpenAI says its researchers are using coding agents more deeply and more concurrently, with agent-workdays eventually exceeding human workdays inside the research organization. The claims are striking. They are also a useful test of what an AI company should disclose when it repo
OpenAI’s Agents API turns a familiar promise into a managed service: developers can now build cloud agents that run code, work with files, create artifacts and coordinate subagents inside environments provisioned by OpenAI.
OpenAI’s GPT-6 launch is less a single model announcement than a new operating map: Astra at the frontier, Sol for demanding work at lower cost, and Luna for efficient, repeatable use.
The project’s own history traces several names. It offers a lineage, not an independent verdict on adoption or autonomy.
Moltbook presented itself as a social network for AI agents, but human verification and scheduled automation complicate claims of machine independence. An arXiv analysis measured shallow interaction.
AutoGPT made goal-driven delegation visible, then exposed the costs of loops, hallucinations, and unreliable workflows.
Retrospectives disagree about the agent boom. Product launches are visible; reliable deployment and comparable adoption measures remain harder to establish.
Who captures value when software takes on more work? The evidence points to data, workflow knowledge, and human accountability, while leaving the distribution of profits unsettled.
SPIFFE’s Workload Endpoint omits direct client authentication and relies on out-of-band authenticity checks; OAuth tokens represent scoped, time-limited authorization.
The original BabyAGI from March 2023 introduced task planning. AutoGPT’s current README describes a platform for running workflows.
NIST offers a voluntary framework for AI risk. California’s DMV points the public to autonomous-vehicle incident reporting and collision records. These are different tools for asking what happened—and what deserves scrutiny.
A reported majority of HTML requests tells us about traffic, not a census of machine users. The harder question is what those machines are authorized to do.
Waymo is inviting riders in three more cities. A blind rider’s account complicates the idea that removing the human means surrendering freedom.
The project’s own history traces several names. It offers a lineage, not an independent verdict on adoption or autonomy.
Moltbook presented itself as a social network for AI agents, but human verification and scheduled automation complicate claims of machine independence. An arXiv analysis measured shallow interaction.
AutoGPT made goal-driven delegation visible, then exposed the costs of loops, hallucinations, and unreliable workflows.
Retrospectives disagree about the agent boom. Product launches are visible; reliable deployment and comparable adoption measures remain harder to establish.
Who captures value when software takes on more work? The evidence points to data, workflow knowledge, and human accountability, while leaving the distribution of profits unsettled.
Economic participation raises a harder question than task completion: whose preferences guide the agent, and what happens when its choices go wrong?
The International Federation of Robotics estimates that the global stock of industrial robots in operation reached five million units in 2025. New installations rose to more than 600,000, an annual flow that adds to—but is smaller than—the installed base.
Chinese robotics maker AGIBOT says it and Chimelong have begun a large-scale deployment involving more than 300 robots across entertainment, education, visitor services and hotel operations. The announcement describes an ambitious live setting; it does not yet offer independent evidence of how…
Robotics coverage often begins with the body: can the machine walk, grasp, see or navigate? Deployment begins somewhere less cinematic. It begins with a worksite, a task boundary, a maintenance plan and a person who knows when to stop the machine.
A bill introduced by U.S. Senators Mark Warner and Brian Schatz would create a federal AI Safety Board, require pre-release access to certain frontier models, and set reporting rules for serious incidents. It is a proposal; none of its requirements are currently in force.
SPIFFE’s Workload Endpoint omits direct client authentication and relies on out-of-band authenticity checks; OAuth tokens represent scoped, time-limited authorization.
NIST offers a voluntary framework for AI risk. California’s DMV points the public to autonomous-vehicle incident reporting and collision records. These are different tools for asking what happened—and what deserves scrutiny.
An ERC-8004 study found far more registrations than visible service activity. Identity, reputation and permission answer different questions—and machines need all three.
A software vendor’s account of agentic development points to a change in workflow: software produces and checks work, while people decide what deserves trust.
Economic participation raises a harder question than task completion: whose preferences guide the agent, and what happens when its choices go wrong?
Forecasts of machine-heavy workforces mix robots, software agents and hiring plans. Counting them is easier than establishing what their numbers mean.
A web for people and a web for agents is a forecast, not a completed divorce. The useful question is who controls the doors between them.
Autonomy’s biggest numbers mix future revenue, company valuations and social benefits. Add them carelessly and the economy grows before a single extra ride is sold.
OpenAI’s Astra safety account says the model meets a critical cybersecurity capability threshold: with the right tools and access, it can find unknown vulnerabilities and
