Mark Cuban: Application-Specific Robots Will Outperform Humanoids
Reactions from ranked influencers
2 postsHumanoids have all kinds of issues. Latency. Processing power. Dexterity. They aren’t designed to be optimal at much of anything. they have to be upgraded continuously. They will be great novelties for a while. “Get me a beer” “ Fold my clothes. “ etc But the downside for humanoids is that application specific devices/robots will be faster/better/cheaper. When they are, our homes and work places will be built and modified to leverage those ASRs (application specific robots) Logistics have already set this path
@mcuban "2 skills AI won’t have for a long long long time. If ever." I think this will occur sooner than you think. This one of the reasons for a full stack connection between robots with sensors that can walk/talk/explore the world and high parameter AI models in the cloud.
In 25 years new construction will be designed to optimize the ASRs ( Application Specific Robots ) available at that time. You don’t think homes will be built and look look the same when unique robots can do a specific physical job faster, cheaper and better ? Won’t happen quickly. But as surely as 3D printing has delivered seawalls (https://www.kinddesigns.com/) faster/better/cheaper , ASRs will take things much further and lead to unique designs and methods
still cannot make heads or tails of Mr. Cuban's opinion, who said robots (plus the AI that allows those robots to work) would take over electrician jobs. Now he says he would rather trust a seeing eye dog ? How is LLM and robotics any different, they are not mutually exclusive. https://twitter.com/mcuban/status/2080355645937926210
Combined views
166.6K
2 posts, first seen 3h ago