OpenAI CEO Sam Altman on potential dangers of AI: "First, we could lose control of the future to AI. The risk is that it moves so fast that people can no longer follow what's happening or intervene when needed."
I asked Opus 5.5 to explain camera focus by building an interactive lens lab
Here's what it came up with after 1 hour 26 minutes in one shot, $25.66 API cost
lens.lab.sael.net
Move the focus ring and you can see the glass elements shift the sharp plane through the scene
first came jev. then laya. then laya-mlx. and now kev.
all system one decision models. all within a span of 3 days. never seen a race this fierce before.
let us breathe, folks. let us breathe.
Google just open-sourced ARTEMIS.
It lets Al agents control a real phone like a human.
You give the agent a task in plain English and it operates the device end to end: sees the screen, finds the elements, taps, swipes, navigates, verifies the outcome, and reports the result.
It connects to your agent over MCP: Claude Code, Codex, Cursor, Windsurf, Antigravity, VS Code, and more.
What Google reports:
β 99%+ task completion on AndroidWorld
β 100+ multi-step tasks
β ~3-5 seconds per step in Flash mode
β Targeting that fuses accessibility, OCR, and vision β Screenshots, execution traces, and diagnostics on every run
β A Python SDK to drop into your test pipelines
Tell Codex to build the APK, install it, log in with a test account, walk the checkout flow, flag any unexpected popups, and return screenshots of the failures, and it drives the device itself.
QA, mobile testing, debugging, automation, all of it moves forward.
Link: github.com/google/artemis
100% open source. Apache 2.0.
It's Monday. A lot changed in agents last week.
I went through everything from xAI, Meta, OpenAI, Anthropic, Google, Apple, OpenClaw, GitHub and Perplexity.
Here's what matters, in one thread: