Skip to main content
  1. Home
  2. Computing
  3. News

Watch this robot get wrecked by a chainsaw … and then adapt

It's kind of unnerving, but also incredibly impressive.

Add as a preferred source on Google
ANY ROBOT, ONE BRAIN

Earlier this week, we shared a video showing a humanoid robot recovering at lightning speed after being kicked to the ground. Its ability to get back up again was truly remarkable and demonstrates some of the great strides being made in robotics in recent times.

Now we have an even more extreme video (top) demonstrating how a robot can recover from something far worse than being knocked over: Having its legs sawed off.

The footage from Pittsburgh-based robotics startup Skild AI shows a robot dog using a specially designed AI brain to adapt to having its legs cut off at the knees. And it copes just fine.

“We built a robot brain that nothing can stop,” Skild AI, which has backing from the likes of Amazon and SoftBank, said in a post on X this week. “Shattered limbs? Jammed motors? If the bot can move, the brain will move it — even if it’s an entirely new robot body.”

Indeed, the video shows the dog using its AI brain to regain mobility in response to a range of disabilities, including a locked knee and faulty front legs (it simply gets up and walks on the other two!). And if you sling a heavy weight over it to make it lose balance, it soon adapts to that, too.

The remarkable robot dog — or more accurately, its AI-powered brain — can even handle having its legs extended by wooden poles, or having wheels attached. Short of a rocket-propelled grenade coming its way, nothing seems to be a problem for it.

Skild AI set about creating the AI brain as it was keen to create robots capable of adapting to potentially fast-changing real-world conditions.

To do this, it trained the AI to control not just one robot, but numerous ones all with different bodies.

“It cannot memorize the solution for one body, it must find a strategy that works across all of them,” Skild AI said in a blog post. “When faced with unpredictable scenarios, the AI can now use the strategies it learnt during training and keep going.”

It said it created “a universe with 100,000 different robots and trained our AI to control them all. After a millennia of simulated time, what emerged was a remarkably resilient, omni-bodied brain. We were often surprised with its ability to adapt to scenarios that were very different from what it saw at training time.”

The AI brain has no idea what robot it’s been placed inside when it’s switched on. So, as you can see from when the robot dog’s front legs are disabled, the AI treats it as a small humanoid robot and quickly learns to walk on two legs. It should be noted, however, that the brain currently needs a little time to adapt, though only “milliseconds to minutes, depending on the severity of the change.”

Skild AI’s adaptable robot brain could be incorporated into robots in a range of scenarios, including in unpredictable, dynamic environments like construction sites, warehouses, or disaster zones where conditions are constantly changing.

Its impressive work could also be used to accelerate innovation by providing a flexible AI platform compatible with many types of robots.

Trevor Mogg
Contributing Editor
Not so many moons ago, Trevor moved from one tea-loving island nation that drives on the left (Britain) to another (Japan)…
YouTube’s AI-powered search is rolling out in the US to find videos based on situations you describe
Ask YouTube can find videos based on the situation or idea you describe
youtube ai search feature

YouTube users in the U.S. are getting a new way to search for videos on the web. The company has started rolling out Ask YouTube, its conversational AI search experience, beyond the Premium-only test announced at Google I/O 2026.

Instead of entering a few keywords and scrolling through a standard list of results, users can ask YouTube a complete question. The feature is designed for broader searches where the exact video, channel, or topic may not already be clear.

Read more
Meta’s detection tool fails to identify photos generated by its own Muse Image AI
Meta has created an invisible watermarking tool called Content Seal that is embedded in all images generated by the Muse Image AI.
Meta AI identification tool.

Earlier this week, Meta announced two new AI products, namely, Muse Image and Muse Video. As the name suggests, these are generative AI tools for making photos and video clips using natural language text prompts. Soon after their rollout commenced, these tools sparked controversy because Meta had automatically opted in Instagram users, allowing others to use their publicly posted media and convert them into remixed AI content. But it appears that Meta courted another loss on its side of the court.

What's the problem?

Read more
Your Google AI Studio apps can finally have polished, presentable web links
AI Studio web apps can now use personalized subdomains
google ai studio logos

Google AI Studio has made building a web app surprisingly easy. You can describe what you want, refine the design through prompts, and publish the result without setting up a traditional development environment. An awkward point of friction comes after deployment, when the finished app still has to live behind a long, forgettable Cloud Run link.

Google is now cleaning up that final step. AI Studio lets you assign a deployed web app a personalized address under the “ai.studio” domain, such as “your-app-name.ai.studio.” A recognizable URL should make the project look more presentable in a portfolio, client demo, social post, or internal project page.

Read more