OpenAI and Meta are moving beyond chatbots that wait for a prompt, but their first always-on agents are arriving with sharply different prices, access rules and privacy settings. An Engadget comparison published October 11 describes OpenAI’s Dots and Meta’s Muse as cloud-based workers that can continue handling tasks after a user leaves.
Meta released Muse on September 8, and OpenAI introduced its always-on agents in ChatGPT about three weeks later, according to Engadget. Both products operate on dedicated cloud computers, connect to outside services and can remain active on scheduled or recurring work. That makes them closer to persistent digital operators than ordinary question-and-answer assistants.
Their practical emphasis differs. Engadget reports that Muse can shop, fill forms, arrange travel, send emails and watch for changes such as price movements. It is available through dedicated apps, the web and WhatsApp, and can connect to services including email and calendars. In Engadget’s hands-on testing, Muse organized saved social-media posts into recipe and restaurant collections.
Dots live inside ChatGPT and run on GPT-6 Astra, the report says. Each Dot receives its own cloud computer, can use connected plugins, retains relevant context and can keep working on ongoing research or recurring assignments. Users may also grant a Dot access to a local computer for files and internet browsing, while OpenAI plans to let customers manage several Dots at once.
Neither system is presented as fully independent. Meta says Muse requests approval before sensitive actions such as sending an email or completing a purchase. OpenAI uses permissions, Custom Rules and Auto-review to govern Dot behavior. Both companies warn that their agents can make mistakes, and external services can impose their own restrictions; Engadget notes that Amazon has already blocked Muse from shopping on its site.
Price may be the clearest dividing line. Muse has a capped free tier, plus Power and Maximum plans priced at $20 and $100 per month. Engadget reports that individual access to Dots currently begins with the $100-per-month Pro 100 plan. Business Premium costs $100 per user per month when billed annually or $125 when billed monthly.
The privacy tradeoffs grow with every connected account. Engadget says Meta does not share conversations and data inside Muse’s secure virtual machine with its advertising systems, and users can opt out of using interactions for model improvement. However, Engadget found that model improvement was enabled by default, along with a default setting that requested approval for only some actions.
OpenAI says activity from Dots on personal plans may be used for model improvement depending on a user’s ChatGPT setting, while Business and Enterprise data is excluded from training by default, according to the report. Proactive Dot research remains read-only until a later action clears its normal permission checks.
Availability is fragmented as well. Engadget reports that Dots are rolling out to eligible Pro users outside the European Economic Area, Switzerland and the United Kingdom. Business Premium users can access them across supported ChatGPT regions, while Enterprise access remains in beta. Muse is currently limited to the United States and Canada.
The comparison reveals an early market divide rather than a settled winner. Meta is lowering the cost of experimentation and distributing Muse across familiar messaging surfaces, while OpenAI is embedding Dots into its existing ChatGPT, Work and Codex environment. In both cases, greater usefulness comes from broader access to email, calendars, files and browser sessions, making permission design as important as raw capability.
Comments
Loading comments…