Anki’s Cozmo, the endearing toy robot with an AI-powered personality, got a big upgrade last week in the form of a new app. And basically, it makes him a little more needy.
“To many of his fans, Cozmo is the ultimate pet robot — you can play games with him, sure, but you can also just hang out with him,” Meghan McDowell, senior producer at Anki, wrote in a blog post. “The update transforms the relationship between Cozmo and his owner to one that is more evocative of the relationship that exists between a beloved pet and its owner.”
It’s a little like Tamagochi: When you update the Cozmo app via the Google Play Store on Android or App Store on iOS, you’ll see new progress meters divided into three categories (or “Needs”): Play, Feed, and Tune Up.
The Play meter fills up as you play games with Cozmo and encourage him interact with Power Cubes, the light-up blocks that ship in the box. “Feeding” Cozmo with the Power Cubes, which involves placing one in front of him as he rolls up to it and pretends to eat it, replenishes the Feed meter. And recalibrating Cozmo by tapping a series of arrows on the screen resets the Tune Up bar.
The meters have a tangible effect on Comzo’s demeanor, McDowell said. As they deplete over time, the less likely Cozmo is to play games, perform tricks, and explore its surroundings. Conversely, the fuller Play, Feed, and Tune Up get, the more bubbly Cozmo acts around familiar faces.
“[Like] a pet, sometimes Cozmo has specific needs that require just a little love and care,” McDowell wrote. “By keeping him tuned-up, energized, and entertained, you develop his skills and abilities, making him a sharper playmate and partner in crime.”
Anki also announced expanded availability of the Cozmo Collector’s Edition. In North America, it launches in Toys R Us, Anki.com, and Best Buy Canada on Sept. 17, and goes on sale later this year in Japan, Germany, U.K., France, and Nordic retailers.
Cozmo’s “personality” update comes on the heels of Code Lab, a Scratch-like visual programming language that puts Cozmo’s motor movements, facial recognition, text-to-speech, and more at your fingertips.
“We’ve been working on Cozmo since 2011 and he still continues to evolve every day,” McDowell wrote. “The more time you spend with Cozmo, managing his needs and helping him explore his abilities, the more he’ll be able to do.”
San Francisco startup Dimensional advertised for people to walk robot dogs through city streets. The New York Times reported that the job could include carrying a roughly 33-pound robot back to the office if it overheated. A lift home is an awkward requirement for a labor-saving machine.
Keeping the hardware running is only one challenge. Anthropic's robotics tests found that language models mostly failed when controlling joints directly, but did better when directing trained controllers. Giving an AI agent a body doesn't teach it to handle ordinary household objects.
Soon, a mattress protector may be more than a bed cover. Australian medtech company Sleeptite and researchers at RMIT University have received a US patent for FlexiSense, a smart bedding technology that tracks body position and posture to help caregivers identify when older residents may be at risk of falling.
The technology is being developed for Sleeptite’s in-bed monitoring system, REMi, and was recently on display at The MedTech Conference in Boston. A peer-reviewed study in Flexible and Printed Electronics says the sensors can also be made at scale without sacrificing performance or durability.
AI companies spend enormous amounts of time convincing us their technology is safe and secure. But what happens when those same companies become targets of increasingly sophisticated AI-powered attacks? Anthropic, the company behind Claude, is now facing that question. According to a Washington Post report, cybersecurity startup Cathedral discovered vulnerabilities that let its researchers access sensitive information in Anthropic's internal systems.
The researchers used AI agents to identify potential entry points in less than 12 hours. Interestingly, they didn't need to crack Claude itself. Instead, they exploited weaknesses in third-party business applications that Anthropic uses for everyday operations. Anthropic's security testing program disclosed the findings, which allows outside researchers to identify and report vulnerabilities.