A new artificial intelligence control system enables soft robotic arms to learn a wide repertoire of motions and tasks once, then adjust to new scenarios on the fly, without needing retraining or sacrificing functionality. 

This breakthrough brings soft robotics closer to human-like adaptability for real-world applications, such as in assistive robotics, rehabilitation robots, and wearable or medical soft robots, by making them more intelligent, versatile, and safe.

The work was led by the Mens, Manus and Machina (M3S) interdisciplinary research group — a play on the Latin MIT motto “mens et manus,” or “mind and hand,” with the addition of “machina” for “machine” — within the Singapore-MIT Alliance for Research and Technology. Co-leading the project are researchers from the National University of Singapore (NUS), alongside collaborators from MIT and Nanyang Technological University in Singapore (NTU Singapore).

Unlike regular robots that move using rigid motors and joints, soft robots are made from flexible materials such as soft rubber and move using special actuators — components that act like artificial muscles to produce physical motion. While their flexibility makes them ideal for delicate or adaptive tasks, controlling soft robots has always been a challenge because their shape changes in unpredictable ways. Real-world environments are often complicated and full of unexpected disturbances, and even small changes in conditions — like a shift in weight, a gust of wind, or a minor hardware fault — can throw off their movements. 

Despite substantial progress in soft robotics, existing approaches often can only achieve one or two of the three capabilities needed for soft robots to operate intelligently in real-world environments: using what they’ve learned from one task to perform a different task, adapting quickly when the situation changes, and guaranteeing that the robot will stay stable and safe while adapting its movements. This lack of adaptability and reliability has been a major barrier to deploying soft robots in real-world applications until now.

In an open-access study titled “A general soft robotic controller inspired by neuronal structural and plastic synapses that adapts to diverse arms, tasks, and perturbations,” published Jan. 6 in Science Advances, the researchers describe how they developed a new AI control system that allows soft robots to adapt across diverse tasks and disturbances. The study takes inspiration from the way the human brain learns and adapts, and was built on extensive research in learning-based robotic control, embodied intelligence, soft robotics, and meta-learning.

The system uses two complementary sets of “synapses” — connections that adjust how the robot moves — working in tandem. The first set, known as “structural synapses”, is trained offline on a variety of foundational movements, such as bending or extending a soft arm smoothly. These form the robot’s built‑in skills and provide a strong, stable foundation. The second set, called “plastic synapses,” continually updates online as the robot operates, fine-tuning the arm’s behavior to respond to what is happening in the moment. A built-in stability measure acts like a safeguard, so even as the robot adjusts during online adaptation, its behavior remains smooth and controlled.

“Soft robots hold immense potential to take on tasks that conventional machines simply cannot, but true adoption requires control systems that are both highly capable and reliably safe. By combining structural learning with real-time adaptiveness, we’ve created a system that can handle the complexity of soft materials in unpredictable environments,” says MIT Professor Daniela Rus, co-lead principal investigator at M3S, director of the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), and co-corresponding author of the paper. “It’s a step closer to a future where versatile soft robots can operate safely and intelligently alongside people — in clinics, factories, or everyday lives.”

“This new AI control system is one of the first general soft-robot controllers that can achieve all three key aspects needed for soft robots to be used in society and various industries. It can apply what it learned offline across different tasks, adapt instantly to new conditions, and remain stable throughout — all within one control framework,” says Associate Professor Zhiqiang Tang, first author and co-corresponding author of the paper who was a postdoc at M3S and at NUS when he carried out the research and is now an associate professor at Southeast University in China (SEU China).

The system supports multiple task types, enabling soft robotic arms to execute trajectory tracking, object placement, and whole-body shape regulation within one unified approach. The method also generalizes across different soft-arm platforms, demonstrating cross-platform applicability. 

The system was tested and validated on two physical platforms — a cable-driven soft arm and a shape-memory-alloy–actuated soft arm — and delivered impressive results. It achieved a 44–55 percent reduction in tracking error under heavy disturbances; over 92 percent shape accuracy under payload changes, airflow disturbances, and actuator failures; and stable performance even when up to half of the actuators failed. 

“This work redefines what’s possible in soft robotics. We’ve shifted the paradigm from task-specific tuning and capabilities toward a truly generalizable framework with human-like intelligence. It is a breakthrough that opens the door to scalable, intelligent soft machines capable of operating in real-world environments,” says Professor Cecilia Laschi, co-corresponding author and principal investigator at M3S, Provost’s Chair Professor in the NUS Department of Mechanical Engineering at the College of Design and Engineering, and director of the NUS Advanced Robotics Centre.

This breakthrough opens doors for more robust soft robotic systems to develop manufacturing, logistics, inspection, and medical robotics without the need for constant reprogramming — reducing downtime and costs. In health care, assistive and rehabilitation devices can automatically tailor their movements to a patient’s changing strength or posture, while wearable or medical soft robots can respond more sensitively to individual needs, improving safety and patient outcomes.

The researchers plan to extend this technology to robotic systems or components that can operate at higher speeds and more complex environments, with potential applications in assistive robotics, medical devices, and industrial soft manipulators, as well as integration into real-world autonomous systems.

The research conducted at SMART was supported by the National Research Foundation Singapore under its Campus for Research Excellence and Technological Enterprise program.

The Department of Homeland Security struck a $1 billion purchasing agreement with Palantir last week, further reinforcing the software company’s role in the federal agency that oversees the nation’s immigration enforcement.

According to contracting documents published last week, the blanket purchase agreement (BPA) awarded “is to provide Palantir commercial software licenses, maintenance, and implementation services department wide.” The agreement simplifies how DHS buys software from Palantir, allowing DHS agencies like Customs and Border Protection (CBP) and Immigration and Customs Enforcement (ICE) to essentially skip the competitive bidding process for new purchases of up to $1 billion in products and services from the company.

Palantir did not immediately respond to a request for comment.

Palantir announced the agreement internally on Friday. It comes as the company is struggling to address growing tensions among staff over its relationship with DHS and ICE. After Minneapolis nurse Alex Pretti was shot and killed in January, Palantir staffers flooded company Slack channels demanding information on how the tech they build empowers US immigration enforcement. Since then, the company has updated its internal wiki, offering few unreported details about its work with ICE, and Palantir CEO Alex Karp recorded a video for employees where he attempted to justify the company’s immigration work, as WIRED reported last week. Throughout a nearly hourlong conversation with Courtney Bowman, Palantir’s global director of privacy and civil liberties engineering, Karp failed to address direct questions about how the company’s tech powers ICE. Instead, he said workers could sign nondisclosure agreements for more detailed information.

Akash Jain, Palantir’s chief technology officer and president of Palantir US Government Partners, which works with US government agencies, acknowledged these concerns in the email announcing the company’s new agreement with DHS. “I recognize that this comes at a time of increased concern, both externally and internally, around our existing work with ICE,” Jain wrote. “While we don’t normally send out updates on new contract vehicles, in this moment it felt especially important to provide context to help inform your understanding of what this means—and what it doesn’t. There will be opportunities we run toward, and others we decline—that discipline is part of what has earned us DHS’s trust.”

In the Friday email, Jain suggests that the five-year agreement could allow the company to expand its reach across DHS into agencies like the US Secret Service (USSS), Federal Emergency Management Administration (FEMA), Transportation Security Administration (TSA), and the Cybersecurity and Infrastructure Security Agency (CISA).

Jain also argued that Palantir’s software could strengthen protections for US citizens. “These protections help enable accountability through strict controls and auditing capabilities, and support adherence to constitutional protections, especially the Fourth Amendment,” Jain wrote. (Palantir’s critics have argued that the company’s tools create a massive surveillance dragnet, which could ultimately harm civil liberties.)

Over the last year, Palantir’s work with ICE has grown tremendously. Last April, WIRED reported that ICE paid Palantir $30 million to build “ImmigrationOS,” which would provide “near real-time visibility” on immigrants self-deporting from the US. Since then, it’s been reported that the company has also developed a new tool called Enhanced Leads Identification & Targeting for Enforcement (ELITE) which creates maps of potential deportation targets, pulling data from DHS and the Department of Health and Human Services (HHS).

Closing his Friday email to staff, Jain suggested that staffers curious about the new DHS agreement come work on it themselves. “As Palantirians, the best way to understand the work is to engage on the work directly. If you are interested in helping shape and deliver the next chapter of Palantir’s work across DHS, please reach out,” Jain wrote to employees, who are sometimes referred to internally as fictional creatures from The Lord of the Rings. “There will be a massive need for committed hobbits to turn this momentum into mission outcomes.”

An Android-specific malware targeting mobile device takeover appears to use generative AI (GenAI) services in its execution flows to maintain persistence on the victim’s smartphone, researchers at ESET have reported.

The raison d’être of the newly-discovered PromptSpy malware is to deploy and run a virtual network computing (VNC) module on the victim’s device, enabling attackers to capture lockscreen data, gather device information, take screenshots and record activity, and block uninstallation.

But to do so it must first establish persistence on the device, and it is here that GenAI comes into play, said the ESET team. They claimed that PromptSpy uses the onboard Google Gemini service to interpret onscreen elements and provide it with dynamic instructions on how to execute a specific gesture that will enable it to remain in the device’s recent app list. This, in theory, stops it being easily swiped away by the user or killed by the system.

ESET researcher Lukáš Štefanko said that while GenAI plays only a minor role in PromptSpy’s execution flow it could have a significant impact on the malware’s potential adaptability.

“Since Android malware often relies on UI-based navigation, leveraging generative AI enables threat actors to adapt to more or less any device, layout, or operation system version, which can greatly increase the pool of potential victims,” he said.

“Even though PromptSpy uses Gemini in just one of its features, it still demonstrates how implementing these tools can make malware more dynamic, giving threat actors ways to automate actions that would normally be more difficult with traditional scripting.”

Štefanko said that based on localisation clues and distribution vectors, PromptSpy seems to be run by a financially-motivated threat actor, exploits Morgan Chase branding, and may primarily target users in Argentina.

However, he also stressed that the malware has not yet popped up in ESET’s wider telemetry, which may suggest it is a proof of concept (PoC) at this point in time. Nor has it been observed on the Google Play store – it can only be downloaded by a dedicated website that its victims would need to be conned into visiting.

Computer Weekly understands that Štefanko’s discovery has been shared with Google via the App Defense Alliance programme, and Android users should already be automatically protected against known versions of it by the Google Play Protect service.

In the unlikely event that PromptSpy has somehow infected their device, victims can remove it by rebooting their phone into Safe Mode, which disables third-party applications and enables them to be uninstalled normally.