Open AI’s Dot is here: a direct competitor to Meta’s Muse! I just gave it a try, and it’s basically like having a personal agent from OpenAI installed within OpenAI itself. It can think autonomously and help you complete and refine a ton of tasks. Here’s the official description: Dot is a highly capable, always-online intelligent agent that understands what matters to you, works tirelessly on your behalf, and takes important tasks off your plate. It even has its own computer that it can control. Of course, if you trust it enough, you can also grant it access to control your computer. The difference from regular ChatGPT is that Dot can carry tasks across conversations without restarting every time. Powered by GPT-6 Astra, Dot connects to 4,000+ apps via plugins. With your authorization, it can also use your computer, learn your preferences, decision-making criteria, and work style from feedback, handle multiple projects simultaneously, and conduct proactive, read-only research in the background. At key moments, it will report back, ask questions, or request your approval. Some official use cases include: tracking client feedback and submitting PRs with tests, updating release materials after scope changes, rerunning analyses after data updates, revising proposals, editing content in your style, and writing posts. Early testers have even said it found a forgotten invoice, prepared the content, and sent it out after confirmation. This update feels like a direct response to Meta’s Muse, which launched earlier this month, as well as other persistent agents like Instinct: all of them are essentially cloud computers + browsers + cross-app automation. For now, I haven’t granted it access to my local computer. It’s using its own cloud computer!