Category: Artificial Intelligence

Posted on

Can Robots and AI Systems Be Creative?

Can robots and AI systems be creative? Since the rise of generative AI, this question has been the subject of intense debate. The issue is not merely whether machines can produce images, texts, or music. It is also about what we understand by creativity and art in the first place, and whether they are necessarily tied to human beings. The Biennale Arte 2026 could have been an excellent place to engage in this debate. In fact, it does so only occasionally. In the Chinese Pavilion, the humanoid robot Astribot S1 writes in the style of calligraphy master Wang Dongling. Visitors can take the works of art, if that is what they are, home with them. Zhang Zhoujie, meanwhile, has computers develop forms for his “digital chairs” based on predefined rules. The designer tries to exercise as little control as possible, giving the computer a degree of creative freedom of its own. In the Venice Pavilion, “Echo/SommersiVo” uses artificial intelligence to translate visitors’ memories and impressions into a constantly changing world of sound and images. Those involved explicitly describe the process as a collaboration between humans and AI in the creative process. These are interesting approaches, but they remain exceptions. At a time when generative AI is challenging our understanding of authorship, originality, and creativity, the world’s most important international exhibition of contemporary art barely addresses this development. The Biennale Arte 2026 did not have to answer the question of whether machines can be creative. But it should have asked it.

Posted on

Robotaxis in Switzerland

Robotaxis have made their way to Switzerland. Since September 28, 2026, self-driving cars have been available for booking through the iamo app in Zurich’s Furttal region. Three electric Nissan Ariyas are operating in Otelfingen, Dänikon, Hüttikon, and Boppelsen. They are equipped with technology from automation specialist WeRide. The vehicles feature six LiDAR sensors, eleven cameras, and eight additional sensors that provide a 360-degree view of their surroundings. The service includes nearly 100 digitally designated pickup and drop-off points. A ride costs a base fare of CHF 4.70 plus CHF 0.38 per kilometer; Swiss Half Fare Cards and GA Travelcards are not accepted. The service operates daily from 1 to 5 p.m. Up to three passengers can ride at a time, accompanied by a safety driver who can intervene in an emergency. Autonomous driving is not entirely new to Switzerland. As early as 2016, PostAuto’s SmartShuttles were operating in Sion in the canton of Valais. The vehicles used at the time, however, operated under very different conditions. Depending on the area, their speed was limited to between 5 and 20 km/h (3 and 12 mph), and they traveled along virtual tracks. The Furttal project now takes a crucial step further: The Nissan Ariyas operate in regular traffic and have to handle far more complex situations than a bus traveling along a predefined route. That is precisely where the pilot project stands to provide valuable insights. And those insights are unlikely to come without incidents. In a real-world trial of this kind, accidents should be expected. It remains to be seen how the public will react.

Posted on

AAAI Announces New Venue

The AAAI Spring Symposium Series fosters an intimate environment for emerging AI communities to engage in workshop-style discussions. After several editions away from a university setting, the series appears to have found a new academic home. The AAAI 2027 Spring Symposium Series will take place from March 22 to March 24, 2027, at Santa Clara University in Santa Clara, California. For many years, Stanford University provided the setting for the Spring Symposia, while more recent editions have taken place at the Hyatt Regency and other hotels at San Francisco Airport. Santa Clara University now provides a new academic setting in Silicon Valley for this established series and its distinctive format of several small symposia running in parallel. At the same time, AAAI has issued the Call for Proposals for the 2027 Spring Symposium Series. Proposals for individual symposia are due November 17, 2026, with notifications scheduled for December 8. AAAI expects to select approximately eight symposia covering a broad range of topics within and around artificial intelligence, with a particular interest in emerging research areas and communities. The move to Santa Clara University thus marks a new chapter for the AAAI Spring Symposium Series and brings the event back to a university campus. The Call for Proposals is available on the Spring Symposium Series site (Photo: Cristiano Tomás, CC BY-SA 4.0).

Posted on

Kissed by the Muse?

According to the article “Tamagotchi trifft KI-Agent” (“Tamagotchi Meets AI Agent”) (Heise, September 24, 2026), Meta has unveiled Muse Charm, a small wearable device designed to serve as a home for its new personal AI agent, Muse. The gadget, which is reminiscent of a Tamagotchi, can similarly be attached to a keychain or carried in a pocket. It features a screen displaying Jolly, Muse’s avatar. As Heise reports, the assistant is activated via a fingerprint sensor, while cameras are intended to allow it to perceive the user’s surroundings. The device is also expected to feature a microphone, two cameras, and 5G connectivity, although the company has yet to release full technical specifications. Mark Zuckerberg said at Meta Connect that Muse Charm is not yet fully finished and that only a small number of units have been produced so far. Nevertheless, according to Heise, Meta plans to launch the device in time for the December holiday shopping season. Pricing and regional availability have yet to be announced. With Muse Charm, Meta is entering the still-emerging field of physical AI, where products such as Humane’s AI Pin and Rabbit’s R1 have so far failed to achieve lasting success. It remains to be seen whether the company has truly been kissed by the Muse.

Posted on

Oliver Bendel Joins the University of Potsdam as Visiting Scholar

Prof. Dr. Oliver Bendel of the FHNW School of Business will be a visiting scholar at the University of Potsdam from September 1, 2026, to August 31, 2027. He considers himself a philosopher of technology, focusing on technology ethics and emerging fields of applied ethics, particularly information ethics, robot ethics, and machine ethics. In information ethics and robot ethics, he takes an empirically informed or empirically grounded approach that remains close to the subject matter. Ethical judgments are preceded by as precise an empirical and technical analysis as possible: The first step is to determine the specific characteristics of a given system, the functions it provides, the data it processes, and the effects it produces. Only on this basis are normative questions concerning issues such as informational autonomy, privacy, responsibility, or well-being examined. Normative claims are not derived from description alone, and empirical findings and normative premises remain clearly distinguishable. In machine ethics, Bendel complements this approach with a constructive-experimental methodology. He develops and tests prototypes himself, including specialized chatbots, voice assistants, and household robots in which moral rules are implemented. Here, construction serves as a method of inquiry: Its initial purpose is to demonstrate whether and how particular moral rules can be operationalized and implemented technically. The resulting artifacts are then systematically examined to reveal the possibilities, limitations, contradictions, and unexpected consequences of their implementation. This creates a feedback loop involving ethical conceptualization, technical construction, empirical testing, and ethical reflection. Overall, this methodology can therefore be characterized as an empirically grounded and constructive-experimental approach to technology ethics and related fields of research. Since 2022, Oliver Bendel has also been an associated researcher with the Potsdam Embodied Cognition Group (PECoG) at the University of Potsdam.

Posted on

Inclusive AI for Wearables

The master’s thesis “Inclusive AI for Wearables: Assistive Glasses as a Case Study” by Myriam Rellstab, a student at the FHNW School of Business, has reached an important milestone. In the summer of 2026, she completed her comprehensive thesis proposal under the supervision of Prof. Dr. Oliver Bendel. The proposal already provides some interesting preliminary findings. Consumer-oriented smart glasses and assistive systems are increasingly based on the same technological foundations, including computer vision, speech recognition, natural language processing, contextual awareness, and real-time information processing. Nevertheless, research and development still tend to treat mainstream and assistive smart glasses as separate categories. This suggests an important opportunity for Inclusive AI: future smart glasses could be conceived as shared technological platforms serving people with and without disabilities rather than as separate mainstream and assistive products. A central question emerging from the proposal concerns the appropriate design strategy. Rellstab distinguishes between universal, specialized, and modular approaches. Universal solutions promise a coherent and inclusive experience but may not adequately address highly specific needs. Specialized solutions can provide targeted support but risk maintaining the separation between mainstream and assistive technologies. Modular systems could offer a middle ground by combining a common platform with components and functions that can be adapted to individual requirements, although this comes with greater complexity. The next stages of the thesis will combine a structured literature review and comparative analysis of existing smart-glasses systems with qualitative interviews involving blind, low-vision, and Deaf or hard-of-hearing users. The ultimate goal is to derive design requirements and develop a conceptual framework for future inclusive AI-supported smart glasses.

Posted on

A Decade Before Article 50: AI Transparency in the European Parliament

Article 50, “Transparency obligations for providers and deployers of certain AI systems,” of the EU Artificial Intelligence Act (EU AI Act) states: “Providers shall ensure that AI systems intended to interact directly with natural persons are designed and developed in such a way that the natural persons concerned are informed that they are interacting with an AI system, unless this is obvious from the point of view of a natural person who is reasonably well-informed, observant and circumspect, taking into account the circumstances and the context of use.” Ten years earlier, the European Parliament had already sought advice from Oliver Bendel on this very topic. In his presentation, “Moral and Immoral Machines”, delivered at the event “Merging of Man and Machine: Questions of Ethics in Dealing with Emerging Technology” in Brussels on September 8, 2016, Bendel introduced GOODBOT, a chatbot he had initiated in 2013 as part of his research on machine ethics. GOODBOT featured several escalation levels and repeatedly made it clear to users that it was merely a machine. Other speakers in the “Ethics & Society” session included Enno Park, Dana Lewis, Yvonne Hofstetter, and Constanze Kurz. At the “Digital Europe” Working Group conference on robotics on November 8, 2017, Bendel also spoke online about related issues in machine ethics. Referring to a care robot, he raised the question: “Should the robot make clear that it’s just a machine?” The transparency obligations set out in Article 50 entered into force on August 2, 2026 (Photo: Sebastian Raible).

Posted on

Euclid and Moira Tell Stories

Humanoid robots are becoming increasingly convincing conversational partners, but can they also become good storytellers? This intriguing question will be explored at New Scientist Live 2026 in London during the Family session “Science Bites: Why are robots bad story tellers?” on October 10 from 12:40 to 1:20 p.m. Computer scientist Dr. Carl Strathearn and lecturer and computer scientist Emilia Sobolewska will present their two AI-powered humanoid robots, Euclid and Moira, and let them engage in a live conversation. While today’s AI systems can generate impressive text, storytelling remains a difficult challenge because compelling narratives require contextual understanding, shared knowledge, creativity, and the ability to connect ideas in meaningful ways. During the demonstration, the two robots will exchange thoughts and ideas, and the audience may even witness them gradually drifting into hallucinations and constructing a shared fantasy world. The researchers hope that these spontaneous interactions could offer new insights into how imagination and creativity might emerge in autonomous AI systems. Euclid has already attracted considerable attention this summer as one of the highlights of the 18th International Conference on Social Robotics (ICSR + Art 2026) in London, where visitors were able to interact directly with the advanced humanoid robot. Developed by Carl Strathearn and Emilia Sobolewska, Euclid combines automatic speech recognition, computer vision, generative AI, skin sensors, and an innovative robotic mouth that produces highly realistic speech movements through machine learning. Unlike many famous humanoid robots with idealized faces, Euclid deliberately represents an older man with deeply lined features, challenging established design conventions while emphasizing authenticity, diversity, and natural human interaction. The upcoming New Scientist Live session now offers another opportunity to experience Euclid in action, this time in an experiment that explores one of the most fascinating frontiers of artificial intelligence: whether machines can move beyond generating language and begin creating genuinely imaginative stories together

Posted on

Terms and Concepts Introduced by Oliver Bendel

Since 2012, Oliver Bendel has introduced or significantly shaped a number of terms and concepts in academic discourse, particularly in the fields of machine ethics, social robotics, and the philosophy of technology. The Munchausen machine (Münchhausen-Maschine) refers to an immoral machine that systematically generates or disseminates falsehoods. The LIEBOT (LÜGENBOT) is a specific implementation of this concept. It is a chatbot conceived by Bendel and realized as a prototype that systematically tells lies. The project was used to explore the capabilities, risks, and ethical implications of deceptive dialogue systems. The reversed cyborg, also referred to as the inverted cyborg, describes the counterpart of the classical cyborg: instead of integrating technical structures into biological ones, it incorporates biological structures into technical ones. Robot enhancement refers to the deliberate extension and improvement of robots by adding sensory, motor, cognitive, or communicative capabilities and constitutes the counterpart to human enhancement. A wearable social robot is a social robot that is worn on the body or attached to clothing and accompanies its user through communication, assistance, or social interaction in everyday life. With “mythomorphic design”, Bendel describes an approach to robot design in which the models are drawn not from humans or animals but from figures found in mythology, fantasy, or science fiction. Animal-Machine Interaction (AMI) is a field of research and application that Bendel has significantly shaped, though he did not coin the term itself. It systematically examines interactions between animals and machines while addressing both technical and ethical issues. Its primary focus is the design and development of machines for interaction with animals rather than the study of animals themselves, extending to the creation of animal-friendly machines. Machine ethics also plays an important role in this field. The proxy machine with proxy morality is a concept in machine ethics according to which a machine acts as a moral proxy by executing moral rules defined by humans, without itself possessing consciousness or responsibility. Other important approaches in this field include annotated decision trees and morality menus (Photo: Jork Weismann).

Posted on

More News of the World

In 1977, Queen released the album “News of the World”. Its cover likely fascinated or disturbed many teenagers and adults at the time, a reaction that later found its way into popular culture – for example, in the animated sitcom “Family Guy”, where Stewie is terrified of it. The artwork depicts a gigantic humanoid robot holding two dead people in its right hand while blood drips from its left. Science fiction artist Frank Kelly Freas had already used a similar motif in one of his earlier works. Brian May has repeatedly emphasized that “News of the World” represented a deliberate stylistic shift for the band – more direct, rawer, and more contemporary. The cover art clearly reflected that new direction. The guitarist has always been as passionate about science fiction as he is about the reality of outer space. In 2007, after a long interruption, he completed his Ph.D. at Imperial College London with a dissertation on interplanetary dust. A few months ago, he was interviewed for the book “RE:VIEW: Wozu die Menschheit retten? (RE:VIEW: Why Save Humanity?)”. The book’s first section features extensive ESA imagery from the AI-supported planetary defense mission Hera, complemented by works of AI-generated art. In the second section, Hera’s own AI system takes on the role of interviewer, conducting “thirty in-depth conversations with figures from the worlds of art, science, philosophy, and society” (publisher’s description, own translation). In addition to Brian May – who was thus able to discuss both science fiction and the reality of space – the interviewees include, among others, the world’s first female space tourist Anousheh Ansari, “Interstellar” visionary Paul Franklin, PETA founder Ingrid Newkirk, former Guns N’ Roses guitarist Ron “Bumblefoot” Thal, science fiction author Andreas Eschbach, artist Jonathan Meese, and technology philosopher and machine ethics expert Oliver Bendel. The book is being published by We Mind Publishing in collaboration with the agency Jung von Matt. It is already listed on Amazon, where extensive information about the book is now available, and is scheduled for release in September 2026.