How OpenAI’s Updated Codex Turns Your Old PCs into a Personal AI Compute Fleet
OpenAI’s latest Codex update adds locked‑screen system control and lets any device with the client join a distributed network, enabling users to repurpose idle computers as a coordinated AI compute fleet while raising security and platform‑compatibility questions.
TestingCatalog reported that OpenAI is expanding Codex’s remote‑control capabilities in two directions: allowing system operations while the host computer is locked and enabling all devices with the Codex client to join a dedicated distributed‑compute network.
The new “Locked Use” setting lets Codex invoke screen actions and run tasks in the background even when the machine is locked or asleep, addressing a limitation that previously required an unlocked screen. Anthropic’s Claude Code offers a similar mobile‑to‑desktop control but is also locked‑screen‑bound.
The cross‑device networking feature lets any installed Codex client—whether an idle Mac Mini at home, a workstation at the office, or an old computer at a relative’s house—to become part of a “Codex network.” A user can issue a natural‑language command from a primary device (phone or main computer) and dispatch different tasks to each node, such as large‑file processing, automation scripts, or model inference, effectively creating a personal AI agent fleet.
For ordinary users this means dust‑gathering hardware can be repurposed without complex remote‑desktop setup; a single command can trigger downloads, video editing, or backups on a home PC. Independent developers and small teams see the possibility of replacing expensive cloud servers with a private cluster of idle machines, reducing cost.
Developers note that the design hits the core pain point of personal compute by eliminating the need for public‑cloud farms and keeping data local.
Some users predict that native account‑level cross‑device control could threaten traditional remote‑access tools such as TeamViewer and networking solutions like Tailscale.
Security concerns are raised about the scope of the Locked Use permissions, per‑device configuration, and the boundaries of cross‑device operations.
Practical questions include whether a 24/7 online node is required, if all devices must share the same ChatGPT account, and whether open‑source alternatives exist.
The feature is still in development with no announced launch date. An open question remains whether Apple will allow bypassing the lock‑screen for system‑level screen control, as this may conflict with macOS security policies.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
AI Engineering
Focused on cutting‑edge product and technology information and practical experience sharing in the AI field (large models, MLOps/LLMOps, AI application development, AI infrastructure).
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
