Connect with us

Technology

What is DAAAM? MIT Develops AI System That Gives Robots Spatial Memory.

The DAAAM AI system developed enables robots to combine spatial mapping with language understanding, remember, describe, and interact with their environments.

Published

on

Humanoid robots working on an automated factory assembly line using AI and robotics technology
Humanoid robots operate on an automated assembly line, showcasing the future of AI-driven manufacturing. Image Credit: iStock/SweetBun Factory

Researchers at the Massachusetts Institute of Technology have introduced DAAAM AI system, a new artificial intelligence system designed to enable robots to better approximate human-like understanding of space, memory, and context. DAAAM stands for Describe Anything, Anywhere, Anytime, at Any Moment. This framework represents a major step toward enabling machines to interpret and remember the world in a meaningful way.

DAAAM AI System Moving Beyond Traditional Mapping

Conventional robotic systems depend largely on spatial maps to understand environments. While these maps are effective for movement, they often lack descriptive depth. A robot may know where something is, but not what it is or how it relates to its surroundings.

DAAAM AI system mapping environment with object detection and language-based descriptions in a 3D spatial model
DAAAM AI system converts spatial data into language-based descriptions, enabling robots to understand and interact with their surroundings. Image credit: MIT News

DAAAM addresses this limitation by combining robotic mapping with advanced computer vision. As a robot explores its environment, it not only builds a three-dimensional map but also attaches detailed descriptions to objects within that space. This allows the system to transform raw spatial data into a structured, information-rich memory.

“If we want robots to work side-by-side with humans and interact better with humans, they must speak the same language. The robot must be able to reason about time and space the same way humans do. That is essentially what our method is doing. It is turning a traditional map into a language-based map that is easier for the robot to think about and access using language,” said Luca Carlone (Associate professor in MIT’s Department of Aeronautics and Astronautics) while addressing the media.

Understanding Space and Time Together

What makes DAAAM particularly significant is its ability to incorporate both spatial and temporal reasoning. The system records observations over time, enabling robots to recall not just locations, but also past interactions.

For instance, instead of simply identifying a workspace, a robot can remember that a specific component was placed there during a previous task. This capability allows it to respond to complex, real-world queries, such as retrieving an item that was last used at a certain time and place.

Efficiency at Scale

One of the key challenges in developing such systems is managing the large volume of visual data encountered in real-world environments. DAAAM overcomes this by selectively processing key visual frames and grouping nearby objects, allowing it to generate detailed descriptions efficiently.

“We annotate every object only once, so our framework can run in very large-scale environments in real time. And by clustering objects into regions, it can answer a wide range of queries about objects and locations in the environment,” one of the lead authors, Gorlo, explains.

With this, the system reduces computational demands while maintaining accuracy. This makes it suitable for real-time operation, even in large and complex environments.

Integrating Language for Interaction

Another defining feature of DAAAM is its integration with LLMs with tool-based retrieval to reduce hallucinations. This enables robots to interpret and respond to natural language queries, making interactions more intuitive for human users.

Rather than requiring precise commands, users can communicate with the system in everyday language. The AI then retrieves relevant information from its structured memory, providing accurate and context-aware responses.

Potential Applications

The implications of this technology extend across multiple domains. In industrial settings, robotic assistants could help workers locate tools or components based on verbal instructions. In augmented reality, similar systems could assist with navigation or highlight anomalies in complex environments.

A Step Toward Human-Like Intelligence

By merging spatial awareness, contextual understanding, and language processing, DAAAM represents a significant advancement in artificial intelligence. It shifts the role of robots from simple task executors to systems capable of reasoning about their surroundings.

As research continues, technologies like DAAAM could redefine how humans and machines collaborate, bringing us closer to a future where robots not only operate in our world but truly understand it.

EP Staff is the editorial team at EdPublica, an independent media organisation focused on science, education, environment and public policy. The team produces evidence-based news, features, explainers and analysis on issues that shape society and everyday life.

Technology

MIT engineers develop light-controlled robot powered by living muscle

MIT engineers have developed a paper-thin soft robot powered by living muscle cells that responds to flashes of light, allowing it to swim and change direction in water.

Jishnu P

Published

on

MIT engineers demonstrate a soft robot powered by living muscle cells responding to blue light.
MIT engineers developed a soft robot that can flap through water in response to flashes of light. Credits: Photo: Melanie Gonick, MIT

MIT soft robot powered by living muscle cells can move through water in response to flashes of light, according to engineers at the Massachusetts Institute of Technology. The robot, described in a study published in Advanced Functional Materials, is the first very thin, two-dimensional muscle-powered robot demonstrated to achieve locomotion, according to the researchers.

The robot is built from a thin film of gelatin methacrylate, or GelMA, which acts as its flexible skeleton. The film forms two fins, each covered with a layer of living muscle cells thinner than a strand of hair.

The cells are genetically engineered to contract when exposed to light. When researchers shine light on one fin, the muscle cells contract and cause the fin to flap, pushing the robot through the water. By changing which fin is illuminated and how frequently the light is flashed, researchers can control the robot’s movement and direction.

In experiments, the robot was able to swim through a simple maze placed inside a dish of water. At its fastest, it travelled a distance of about four times its body length in one minute.

While that speed is slow compared with human swimmers, the experiment demonstrates that even a single layer of muscle cells can generate enough force to move a small robot through water.

“It takes a lot of force to move through water versus air. The robot’s quite strong, given its size.” said in a media statement issued by MIT.

MIT soft robot

Tissue-engineered soft-robotic ray. Representational image .Credit: Karaghen Hudson and Michael Rosnach

MIT Soft Robot: Designing the robot around its muscle

The research builds on an earlier project by Raman’s group that used muscle cells to create an artificial structure inspired by the human iris.

In that work, researchers grew muscle cells on a thin gel containing a pattern of grooves. When exposed to light, the cells contracted in different directions, producing movement similar to the way the iris expands and contracts.

“People hadn’t seen this muscle architecture engineered from scratch before ,and the cells were moving in multiple directions. But they only moved about 100 microns. From a robotics perspective, their movements were tiny.” MIT engineer said. 

They therefore focused on the material supporting the muscle cells. The team tested gels with different levels of stiffness, compositions, thicknesses and groove patterns.

The researchers found that square-bottomed grooves helped the muscle cells align more effectively than curved grooves. Better alignment allowed the cells to fuse into stronger muscle fibres and generate greater force.

The team also replaced fibrin, an ultrasoft gel used in the earlier design, with GelMA. Stiffer GelMA provided better support for the cells and allowed them to produce more force.

A GelMA film about half a millimetre thick was found to provide enough support while remaining light enough for the muscle cells to contract without pulling away from the surface.

The researchers also “trained” the muscle cells by repeatedly exposing them to flashes of light before incorporating them into the robot.

The resulting device effectively has two independent muscle-powered fins. Lighting one fin causes movement on that side, while illuminating both makes both fins flap.

The researchers say such biohybrid robots could eventually have applications in aquatic environmental monitoring. Because they combine living tissue with engineered materials, the technology could potentially be adapted for delicate environments where conventional rigid robots may be difficult to use.The MIT soft robot shows how living muscle cells can power biohybrid machines. The use of light-controlled robots could eventually be used for environmental monitoring and other delicate tasks in aquatic environments where conventional robots may be difficult to operate.

For now, the MIT team is working to improve the robot’s body design and increase its swimming speed to further develop the experiment.

Continue Reading

Technology

AI Is Growing Fast, But Women Remain Underrepresented

Men hold nearly 80% of AI jobs and 90% of senior leadership positions, while 44% of 133 AI systems showed gender bias, UN Women reports.

Published

on

Girl standing beside a humanoid robot, illustrating the gender gap in artificial intelligence and who shapes the technology
A girl stands beside a humanoid robot, highlighting the gender gap in the people developing and shaping artificial intelligence. Representational image. Image credit: Pavel Danilyuk/Pexels

Artificial intelligence is spreading across workplaces, public services and decision-making systems, but women remain underrepresented widening the gender gap, in the workforce building and governing the technology. Men hold nearly eight in 10 AI jobs and almost nine in 10 senior leadership positions shaping the technology, according to UN Women. The figures were released as the agency launched its Digital, Innovation and AI Hub during the 81st session of the United Nations General Assembly in New York.

The imbalance extends beyond employment. A global analysis of 133 AI systems found that 44 per cent showed gender bias, according to UN Women. The finding raises concerns about how systems developed with incomplete representation may perform when used across populations with different experiences and needs.

Women are also facing risks from AI-generated and AI-assisted abuse. A UN Women survey of women human rights defenders, activists and journalists found that nearly one in four respondents had experienced AI-assisted online violence. The survey covered 641 women across 119 countries.

AI-generated images, manipulated content and deepfakes have added new tools to online harassment. For women whose work places them in the public sphere, the effects can extend beyond social media.

Gender Gap: Women Remain a Minority in AI

The gender imbalance starts well before senior leadership. Women accounted for about 30 per cent of people working in AI in 2024, according to data cited by UN Women from the World Economic Forum. Women are also underrepresented in STEM education and careers in many parts of the world. This limits the number of women entering the teams that develop AI systems, conduct research, build datasets and make decisions about how the technology is deployed.

The problem is not restricted to technical roles. AI increasingly affects recruitment, healthcare, education, financial services and public administration. Decisions about the use of these systems can therefore affect people who have no role in developing them. Bias can enter at several stages, including through training data, system design, testing and deployment. Underrepresentation in datasets can also leave certain experiences poorly reflected in the resulting systems.

UN Women has called for engineers, social scientists, human-rights specialists and gender experts to be involved in technology development and oversight. The issue becomes particularly important when AI is used for decisions involving access to jobs, services or other opportunities.

AI Used in Online Violence

The technology has created additional ways to target women in public life. Among the 641 women surveyed by UN Women, nearly a quarter reported experiencing violence assisted by AI. The respondents included journalists, activists and human rights defenders working across 119 countries.

AI can be used to produce or manipulate images, videos and other material used for harassment. Such attacks can damage professional reputations and discourage women from participating in public debate. Research cited by UN Women found that the share of women journalists reporting that online violence had spilled into the physical world increased from 20 per cent in 2020 to 42 per cent in 2025.

The Access Gap Starts Earlier

The shortage of women in AI is connected to unequal access to digital technologies and STEM education. Women in low-income countries continue to have lower internet access than men, according to UN Women. Differences in access to digital tools and skills can affect who enters technology-related education and employment.

There is also an economic cost to the divide. The UN Women–UN DESA Gender Snapshot 2025 estimated that closing the global gender digital divide could generate $1.5 trillion in additional global GDP by 2030.

The same analysis estimated that closing the divide could benefit 343.5 million women and girls and help lift 30 million women out of poverty by 2050.

UN Women Launches AI and Digital Policy Hub

UN Women has launched the Digital, Innovation and AI Hub to address these gaps. The hub will bring together researchers, governments, technology companies, AI specialists and women’s rights organisations. Its work will include studying how emerging technologies affect women and girls and supporting governments in developing gender-responsive digital and AI policies.

It will also work with technology companies on incorporating gender equality into the design and governance of AI systems.
At the UN General Assembly, the issue sits within a wider debate over AI’s role in employment, economic development, public services, warfare and international security.

For women, the issue extends from access to AI-related jobs to participation in decisions about how the technology is built, regulated and used.

Continue Reading

Society

Digital Detox: Why Taking a Break From Screens Matters

A digital detox can help children and adults reduce screen dependence, reconnect with nature and relationships, and create space for reflection and creativity.

Anoop Krishnan H

Published

on

digital detox
Image credit: Darina Belonogova/Pexels

A digital holiday can offer a practical pause from screens and constant connectivity. From children to working professionals, taking regular time offline can help rebuild attention, creativity, relationships and a healthier balance with technology. A digital detox can help children and adults reduce screen dependence, reconnect with nature and relationships, and create space for reflection and creativity.

Imagine a day without digital devices. Those of us who grew up in the 1990s remember the shift firsthand — from writing letters with ink pens to typing messages on social media and making video calls. Artificial intelligence and rapid technological change now touch nearly every part of daily life, and an internet-first era has drawn humanity into a globally connected network. We ask AI chatbots for advice on everything from recipes to relationships. Yet the love of books and literature hasn’t disappeared — it has simply changed form. Audiobook platforms have grown fast, gaining listeners who once preferred print. At the same time, attention spans are shrinking as short-form video reshapes how we consume information. In an era built around likes, shares and instant search results, there is a real case for finding a better balance between online and offline living.

Children under 16 in particular need more exposure to offline living, and less dependence on screens. Time away from devices helps children build social skills, sharpen critical thinking, and learn to approach problems from multiple angles — all of which support holistic personal development.

Of course, context matters. During the Covid-19 pandemic, online education became the only option once lockdowns were imposed, and digital learning kept formal education running when nothing else could. But in a post-pandemic world, governments are increasingly reconsidering how much unsupervised screen time is appropriate for children. China’s “minor mode” framework restricts screen time by age; the United Kingdom has moved to ban social media for under-16s from 2027; and New Zealand has introduced legislation to do the same. In India, Karnataka announced in its 2026 state budget that it would ban social media use for under-16s, and Goa’s government has said it is studying a similar move. The details of enforcement remain unsettled in most of these cases, but the direction of the debate is clear: policymakers across the world are actively discussing how to limit children’s social media access. A middle path — rather than an outright ban — is worth considering.

Digital Detox Awareness

Schools are well placed to lead here. A monthly digital detox awareness session, run by trained resource persons and built around hands-on, creative activities, could help draw out children’s imagination while gently reducing screen dependence. Students could keep a diary of their experience — what they noticed, what they missed, what surprised them — during each digital detox day. Over time, schools could even form “digital holiday clubs” to mark one day a month as a shared offline day. Practised consistently through school life, this could help a generation grow into adults with more clarity of thought and purpose — provided they use that offline time for something creative and productive, rather than simply waiting it out.

In practice, a life entirely without the internet isn’t realistic for most of us. But digital minimalism is achievable, and a single digital holiday once a month is a reasonable place to start. Switching off completely for one day can open space for new ideas and reconnect us with the natural world.

That day can also be a chance for self-reflection — a deliberate pause to look inward. It can be used to build a skill: writing, cooking, dancing, whatever draws you. It’s an opportunity for offline meetups with friends and family, for cycling a short distance, for reading a book purely because you chose it, not because an algorithm suggested it. A digital holiday can help you rediscover what actually matters to you and reset your priorities. It also strengthens real relationships — the kind built through presence, not notifications — and leaves room for practices like yoga and meditation that support genuine mental peace.

Digital Detox Is Harder for Working Professionals

For working professionals, this is harder. Most of us are running behind deadlines, structuring our days around work and family obligations already. Stepping outside that loop, even for a day, takes real intention. But the practice of digital detox is worth the friction — it teaches delayed gratification and reintroduces us to the quieter pleasures of offline living.

None of this is a case against technology. Instant messaging and the broader digital revolution have made services faster and more accessible than ever, and that’s worth acknowledging. But speed and convenience come with a cost if we let them: information overload, and an over-reliance on AI chatbots for decisions that deserve real human judgement. Blindly following AI-generated advice isn’t something to encourage. The internet is a necessity now — but that makes the case for balance stronger, not weaker.

Reconnect with nature. Spend real time with the people who matter to you. And once in a while, take the leap: switch off for a day, and notice the difference it makes.

Continue Reading

Trending