This Ping Pong Robot Can Beat Elite Human Players
Sony’s AI division has designed a robot that can beat elite human players at table tennis. From the paper:
Evaluated in matches against elite and professional players under official competition rules, Ace achieved several victories and demonstrated consistent returns of high-speed, high-spin shots. These results highlight the potential of physical AI agents to perform complex, real-time interactive tasks, suggesting broader applications in domains requiring fast, precise human–robot interaction.
Ace is a fine name, but I might have gone with something like WALL-E Supreme instead. (Robbie Supreme?)
Comments 4
Good for the robot, I guess.
Uhhhhh, so what happens when the paddle is a gun?
This is all very well as a technological demonstration. But in terms of anything else it is as interesting as demonstrating that a sports car can beat a human in a marathon. Yes, it can: so what? Table-tennis is a game played between humans, marathons are run by humans, chess is a game played by humans. Art is something made by humans. I mean, we go to great lengths to prevent humans enhancing their performance in sport: this is just performance enhancement taken to the limit.
We've forgotten that humans matter, somewhere along the line.
I get people having feels about this, I felt similar the first time I saw the product-ized tennis-playing robot (do want!) or the basketball hoop in a youtube video that was rigged up to make every arbitrary throw a swish. But listen to the creator and they’re legit concerned the thing will break itself to pieces, I’d be happy if it got advanced and cheap (that vision tech has to be $$$) and scaled down to do more than help the world’s best train, and could potentially just idly keep interested people amused. Just like the graphic overlay of live fencing hitting social media this week, the spin/axis visualization and seeing path predictions is swank, I’m not inclined to think this isn’t just a net plus but on a trajectory to get holodeck cool. Like how chess players can study openings and scenarios.
I’m a bad competitor (ask my significant other about the various board and video games we’ve tried to have fun with over the years) so I’d rather ‘do karaoke’ or carnival games where you hang out with humans and show off or alternately be embarrassed punching the speed bag. I also appreciate researchers correctly calling out when they’re using machine learning, the worthwhile science that gets slathered over with inaccurate naming like artificial intelligence when you just wrote your domain-specific application to wrap it - yes it can account for discrepancies not in the model, data science is an inspiring field. Pardon the enthusiasm 😅
Hello! In order to comment or fave, you need to be a current kottke.org member. If you'd like to sign up for a membership to support the site and join the conversation, you can explore your options here.
Existing members can sign in here. If you're a former member, you can renew your membership.
Note: If you are a member and tried to log in, it didn't work, and now you're stuck in a neverending login loop of death, try disabling any ad blockers or extensions. Or try logging out and then back in. Still having trouble? Email me!
In order to comment or fave, you need to be a current kottke.org member. Check out your options for renewal.
If you feel like this comment goes against the grain of the community guidelines or is otherwise inappropriate, please let me know and I will take a look at it.
Hello! In order to leave a comment, you need to be a current kottke.org member. If you'd like to sign up for a membership to support the site and join the conversation, you can explore your options here.
Existing members can sign in here. If you're a former member, you can renew your membership.
Note: If you are a member and tried to log in, it didn't work, and now you're stuck in a neverending login loop of death, try disabling any ad blockers or extensions. Or try logging out and then back in. Still having trouble? Email me!