Open-source projects replicate OpenAI’s persistent agents as users report glitches
OpenAI’s recently launched Dots, described as always-on AI agents that work in a cloud computer, have prompted open-source alternatives despite reports of technical problems. The developments highlight interest in persistent AI agents, while raising questions about how reliable they are and whether users want them.
Users have reported failed tasks, connection problems, disappearing agents and refusals; OpenAI did not respond to a request for comment. Tawkit’s OpenDots is a self-hosted template, while Open Dot is a desktop app that can use OpenRouter models and connect with services including Gmail, Slack and GitHub. Its creator, Prathit Joshi, says he uses it to organise tasks and summarise social media posts; open-source options still carry model costs, though they may offer more choice and control.
- OpenAI Dots users report failures and connectivity issues.
- Open-source alternatives offer self-hosting and model choice.
- Demand for always-on AI agents remains uncertain.
New here? Start with this
Artificial intelligence tools that work continuously without needing you to prompt them each time are an emerging technology. OpenAI recently launched Dots, AI assistants that stay active in cloud computers and handle tasks automatically, much like having a dedicated helper working round the clock. This always-on approach to AI has sparked significant interest in what the technology might accomplish for ongoing work.
However, OpenAI's launch has encountered problems, with users reporting incomplete tasks, connection failures and disappearing assistants. Rather than deterring interest in the concept, these technical difficulties have prompted independent developers to create open-source alternatives that users can run on their own computers or servers instead.
These alternatives give users more control and the ability to connect their AI tools with services they already use, such as email, messaging platforms and coding tools. Whilst running these alternatives still involves costs, they appeal to users who want greater choice over how their AI assistants operate and potentially more reliability than the official version.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
Advocates contend that persistent AI agents represent a promising technological development, with current technical problems being expected teething issues in any emerging system. The enthusiasm among open-source developers and real-world applications—such as automating routine tasks and aggregating information—demonstrate genuine user interest. Technical glitches are solvable engineering challenges rather than fundamental flaws, and the existence of both proprietary and self-hosted options provides users with meaningful choice about how they adopt this technology.
The case against
Critics argue that reported reliability problems—including failed tasks, connection failures, and disappearing agents—suggest these systems are not yet dependable enough for users who need consistent performance in their daily workflows. They question whether most people genuinely require always-on autonomous agents monitoring their accounts and data, and raise unresolved concerns about privacy, security, and accountability when systems operate without direct human oversight. The resource costs may outweigh benefits for typical users, particularly when the technology's core value proposition remains unproven.
Read the full article at the source →
Originally published by The Register as “OpenAI dots inspire open source imitators amid technical difficulties”.