Language models (LMs) have already demonstrated remarkable abilities in understanding and generating both natural and formal language. Despite these advances, their integration with real-world environments such as large-scale knowledge bases (KBs) remains an underdeveloped area, affecting applications such as semantic parsing and indulging in "hallucinated" information. This paper is an experimental investigation aimed at uncovering the robustness challenges that LMs encounter when tasked with knowledge base question answering (KBQA). The investigation covers scenarios with inconsistent data distribution between training and inference, such as generalization to unseen domains, adaptation to various language variations, and transferability across different datasets. Our comprehensive experiments reveal that even when employed with our proposed data augmentation techniques, advanced small and large language models exhibit poor performance in various dimensions. While the LM is a promising technology, the robustness of the current form in dealing with complex environments is fragile and of limited practicality because of the data distribution issue. This calls for future research on data collection and LM learning paradims.

Videogames developed in the 1970s and 1980s were modest programs created in a couple of months by a single person, who played the roles of designer, artist and programmer. Since then, videogames have evolved to become a multi-million dollar industry. Today, AAA game development involves hundreds of people working together over several years. Management and engineering requirements have changed at the same pace. Although many of the processes have been adapted over time, this is not quite true for quality assurance tasks, which are still done mainly manually by human beta testers due to the specific peculiarities of videogames. This paper presents an approach to automate this beta testing.

Is the world's favourite cartoon dog about to get her own video game? Online sleuths have discovered a Bluey game may be in the works, after a Twitter bot devoted to Australian video game classification decisions tweeted a new rating: Bluey: the Videogame received a G for General. According to a full listing on the Australian government's classification board website, the game – which has not been announced – originates in the UK, where all commercial rights to Bluey are held by the BBC. And while receiving a classification does not mean the title will be made, or has even been officially licensed, this particular game seems to have heft behind it: it lists Outright Games as its publisher, a British company known for its Paw Patrol and Peppa Pig titles. Bluey: the Videogame is described as a treasure hunt-style game that will be available on PC, PlayStations 4 and 5, Nintendo Switch and Xbox.

The history of decision-making has always been intrinsically tied to the history of technology. Charts and compasses have guided explorers for centuries, and a level is an indispensable instrument for construction workers. New tools allow us to make more informed choices which, in turn, may positively impact technological advancements. This dependence suggests that a change in the technological landscape will have implications in how we make decisions. The last half-century has seen one of the most radical revolutions: the emergence of artificial intelligence (AI), powered by the ever-increasing data we gather.

A global audience of a half-billion gamers have gotten to know the axolotl, which largely cluster in the canals around Mexico City and look like little dragons with a goofy smile. The videogame "Fortnite" trotted out axolotl characters in 2020, and "Minecraft" followed suit last summer. Roblox, a platform with millions of user-made games, has dozens of axolotl-centric ones, including "Axolotl Tycoon" and "Axolotl Paradise." Axolotls appear in "Adopt Me!," one of the most-played games on Roblox. All of the exposure has spawned axolotl memes, YouTube videos, coloring books and nonfungible tokens.

For the new sequel, "The Matrix Resurrections," filmmakers deployed much-higher-caliber technologies, including three-dimensional imagery made using artificial intelligence. But after 22 years of digital evolution, high-end movie effects are approaching a plateau near perfection. "We went from pulling off what seemed to be impossible, to a sort of inability to create surprise" in the movie industry, says John Gaeta, who helped craft the bullet-time effect. He was a visual-effects designer on the first three "Matrix" films; now he is making things for the metaverse. This year the movies presented us with a car slingshotting from cliff to cliff ("F9"); Ryan Reynolds running amok inside a videogame ("Free Guy"); and giant monsters crushing the Hong Kong skyline ("Godzilla vs. Kong").

The UN is calling for a moratorium on artificial intelligence systems "that pose a serious risk to human rights" until research and regulation has been done. It published a report today after concerns that countries and businesses are adopting AI without proper diligence. High commissioner for human rights Michelle Bachelet said that AI can be a "force for good" but stressed that it can still have a profoundly negative, "even catastrophic" effect if used without consideration.The report analyses the ways AI can affect human rights, including privacy, health, and education as well as freedom of movement, expression, and assembly. "Artificial intelligence now reaches into almost every corner of our physical and mental lives and even emotional states," Bachelet writes. "AI systems are used to determine who gets public services, decide who has a chance to be recruited for a job, and of course they affect what information people see and can share online."Bachelet's report says that because of its rapid growth, finding out how AI collects, stores, and uses data is "one of the most urgent human rights questions we face.""The risk of discrimination linked to AI-driven decisions—decisions that can change, define, or damage human lives—is all too real," the report continues. "This is why there needs to be systematic assessment and monitoring of the effects of AI systems to identify and mitigate human rights risks."The UN also calls for significantly more transparency from companies and countries that develop and use AI systems. It's important to note that the UN is not calling for an outright ban—no one got spooked by their latest viewing of The Terminator—just regulation and greater transparency. Bachelet says, "We cannot afford to continue playing catch-up regarding AI—allowing its use with limited or no boundaries or oversight, and dealing with the almost inevitable human rights consequences after the fact. The power of AI to serve people is undeniable, but so is AI's ability to feed human rights violations at an enormous scale with virtually no visibility. Action is needed now to put human rights guardrails on the use of AI, for the good of all of us."You can read the press release and full report on the UN's website.What effect this could have on videogames and similar technology is unclear, though the UN is obviously not talking about AI machine learning like Nvidia's DLSS or AI upscaling tech. One possible way this could affect videogames is if a company develops an AI system that learns how specific people play games and then use that data to present targeted microtransactions, ads, or other things to encourage you to spend money—similar to the method Activision patented in 2017. Whether the UN would consider that a violation of your rights is not something I can answer, but in the end, the report's call for regulation probably doesn't impact videogames in a meaningful way for players—even if you wish it would call to improve the AI in games like Cyberpunk 2077. 

From “penny dreadfuls” to comic books to music, often the best way to make a business boom has been to try limiting it.

What's the first question parents ask when their kid comes begging to play a violent shooter-type game? Kids are often good at making strong arguments for why they should be allowed to play non-gory but murderous games like "Fortnite" and "Among Us." Even in "Minecraft," there are skeletons, zombies and pillagers to kill. Kids will say they understand the difference between games and real life. And besides, they'll argue, these popular games are cartoonish--cute, even.

Jensen Huang: This is the autonomous vehicle computer and all of the software and the infrastructure that goes along with it. We announced a partnership that relates to every single Mercedes-Benz car starting in 2024. We're going to build one architecture that spans the entire fleet. We are going to develop applications and then offer them to their customers and share the economics. Whereas a chip could go for a couple of hundred dollars, autonomous driving applications could go for several thousand dollars.