The word “metaverse” has been used so often in recent years that it can feel more like a buzzword than a concrete idea. Tech companies promise it will change how we work, learn, play, and even relate to one another. Critics worry it will become an escape from reality, a new form of surveillance, or just another way for a few powerful companies to dominate our digital lives. But what does research actually tell us about this emerging concept, beyond the hype?
A recent multidisciplinary study by a large group of researchers from different fields examines this question. They look at the metaverse from many angles: technology, psychology, education, health, the future of work, democracy, public opinion, and law, with the goal to understand what it would really mean for society if it were to become widely adopted.
A full metaverse, as they define it, does not exist yet, but we do have early, partial examples. Online games, augmented reality apps, and social VR platforms give us glimpses of what might come, but they are not yet connected into a single, large-scale, persistent system. The metaverse is defined as a set of shared, three-dimensional spaces that exist in real time and continue to exist even when you log off. In these spaces, digital worlds and digitally enhanced versions of the physical world come together. It’s different from looking at a screen: people appear in these environments, usually as avatars, and can interact with places, objects, and other people in a way that feels similar to being there in person. Ideally, people and their digital belongings could move across different virtual worlds and platforms in the same way that websites are connected today. Thousands or millions of people could, in principle, be connected to the same overall system.
The physical devices that give access to the metaverse are evolving quickly. Virtual reality headsets now commonly track the position and rotation of the head very accurately, often using cameras on the device itself rather than external sensors. This allows the virtual scene to update in a stable way as you move, which is essential for the feeling of presence and for avoiding motion sickness. Artificial intelligence, and especially recent advances in large language models and generative AI, are seen as a key driver for making the metaverse practical. Building detailed, interactive 3D environments by hand is labor-intensive and requires specialized skills: to reach the scale and variety that a metaverse implies, content creation must become as easy as taking a photo or writing a short text. Generative models that can turn text prompts into images, 3D shapes, animations, and even entire scenes are a step in this direction. Other techniques can reconstruct 3D environments from ordinary photographs or videos, making it possible for non-experts to scan real places and bring them into virtual worlds.
On top of that, AI systems can help connect natural language instructions to interactive behaviors by generating code that defines how objects and characters should react. This could allow people to “program” parts of the metaverse simply by describing what they want to happen. However, these technologies still face serious limitations: they require large amounts of computation and energy, their results are not always predictable or controllable, and full end-to-end workflows from prompt to polished interactive experience are still at an early stage. Another domain where AI is likely to be central is that of virtual humans: digital characters that can talk, move, and respond socially in a convincing way. These might serve as tutors, guides, customer service agents or companions.
Because the metaverse is not only a technological project but also a social one, much of the discussion should be focused on potential uses and impacts in different domains of life. In education, there’s significant promise: immersive environments can let students experience historical events from a first-person perspective, practice complex procedures safely, or collaborate with classmates and teachers in shared virtual spaces. Professional training, such as for doctors, engineers, or emergency responders, can benefit from realistic simulations that would be too costly, dangerous, or impossible in the real world. In healthcare, the metaverse could reshape how services are delivered and experienced: virtual consultations, rehabilitation exercises, and group therapies could take place in immersive settings rather than on flat screens. For example, people suffering from anxiety or phobias could undergo exposure therapy in controlled virtual scenarios, gradually facing their fears with the support of a clinician. Social spaces in the metaverse could help reduce loneliness or isolation for people who cannot easily leave their homes. However, these possibilities also raise concerns about unequal access, data privacy, and clinical safety.
While these potential innovations would be invaluable, the dangers of surveillance and manipulation in such rich environments must be acknowledged: Because metaverse platforms will likely track not only what people click or watch, but also how they move, where they look, how long they pay attention, and even physiological signals, they could collect extremely detailed profiles of individuals. These profiles, if misused, could be used to influence behavior in subtle and powerful ways, whether for commercial or political purposes.
The authors also consider how the metaverse could affect democracy and civic life. On the positive side, it could offer new spaces for public debate, citizen assemblies, political education, and cross-border dialogue. People who have difficulty participating in physical events, for reasons of mobility, geography, or safety, might join through avatars. Immersive experiences could make abstract political issues more concrete and emotionally engaging. On the negative side, the same features that foster engagement could also amplify polarization, misinformation, and extremist recruitment. Immersive “echo chambers” might reinforce people’s views and emotions even more strongly than text-based social media, and manipulative experiences could be designed to target specific groups.
Another risk that deserves special attention is that of confusing virtual and physical realities in harmful ways. People could treat virtual objects as if they were physically present, for example, attempting to sit on a virtual chair that does not exist in the real world, with obvious safety implications. Negative group interactions could also lead to increased conflict, and in some cases, playing using an avatar that represents a minority member may lead to greater bias towards that minority.
Despite these concerns, the metaverse isn’t presented as inherently good or bad; instead, it may be a high-stakes transformation that could amplify both positive and negative aspects of our digital society. If designed and governed wisely, it could support more engaging learning, more accessible healthcare, new forms of creativity and social connection, and richer participation in public life. If left to develop without careful oversight, it could increase inequality, enable new forms of exploitation, and erode trust and well-being. The main message of the study is that now is the time for serious, interdisciplinary work on the metaverse: technical research needs to go hand in hand with social science, ethics, law, and public debate, long-term studies of real use, beyond short-term novelty, are necessary to understand how people actually live in these spaces. By looking closely at both the opportunities and the challenges, the study invites us to think of the metaverse not just as an inevitable product of technology, but as a collective project that we can still shape.
