1. I don't think it'll get acquired by a frontier lab. Frontier labs already focus on robotics with enough effort that acquiring pure-software research IP from a company like this is not very useful. OpenAI has way more compute and researchers than these guys, and there likely isn't very much dream is doing on the model side thats different from OpenAI, or what they are capable of doing.

2. However to argue against basically any other robotics startup, I think they would only be acquired if it was a robotics SW company _with real customer distribution_/revenue, or a robotics company focusing on HW that is worth acquiring the IP for. The problem with the latter is that it is also a huge cost, and that cost is only worth it if the ML side actually catches up to the HW. I don't think a frontier lab would want to sustain a full physical supply chain for a non-revenue additive product. And frankly, most robotics companies have no meaningful distribution/revenue or SW/HW IP that is worth acquiring, considering the cost.