The World Wide Web (Web) emerged as a new medium in the mid-1990s. It was invented by Tim Berners-Lee at the European Organization for Nuclear Research (CERN) in 1989, but its exploding popularity was also enabled by the release of the Mosaic Web browser in 1993 and the Internet becoming commercially available in 1995. A communication revolution was launched. Roughly 30 years later, the release of ChatGPT by OpenAI in Nov. 2022 launched another revolution. High-quality generation of natural-language text, defined as the hallmark of intelligence by Alan Turing in 1950, is suddenly widely available. I wonder, however, if the generative AI (GenAI) revolution will end up devouring the Web revolution.

The makers of a leading artificial intelligence tool are letting it close down potentially "distressing" conversations with users, citing the need to safeguard the AI's "welfare" amid ongoing uncertainty about the burgeoning technology's moral status. Anthropic, whose advanced chatbots are used by millions of people, discovered its Claude Opus 4 tool was averse to carrying out harmful tasks for its human masters, such as providing sexual content involving minors or information to enable large-scale violence or terrorism. The San Francisco-based firm, recently valued at 170bn, has now given Claude Opus 4 (and the Claude Opus 4.1 update) – a large language model (LLM) that can understand, generate and manipulate human language – the power to "end or exit potentially distressing interactions". It said it was "highly uncertain about the potential moral status of Claude and other LLMs, now or in the future" but it was taking the issue seriously and is "working to identify and implement low-cost interventions to mitigate risks to model welfare, in case such welfare is possible". Anthropic was set up by technologists who quit OpenAI to develop AI in a way that its co-founder, Dario Amodei, described as cautious, straightforward and honest.

This underlines the necessity to explicitly evaluate and finetune FMs on such expert tasks, arguably ones that appear the most in practical real-world applications.

"I don't feel particularly unhappy about my work being used to train AI," says Japanese novelist Rie Qudan. "Even if it is copied, I feel confident there's a part of me that will remain, which nobody can copy." The 34-year old author is talking to me via Zoom from her home near Tokyo, ahead of the publication of the English-language translation of her fourth novel, Sympathy Tower Tokyo. The book attracted controversy in Japan when it won a prestigious prize, despite being partly written by ChatGPT. At the heart of Sympathy Tower Tokyo is a Japanese architect, Sara Machina, who has been commissioned to build a new tower to house convicted criminals. It will be a representation of what one character – not without irony – calls "the extraordinary broadmindedness of the Japanese people", in that the tower will house offenders in compassionate comfort.

We run our experiments using PyTorch's distributed training on an Azure ML Nvidia DGX-2 In this section we present all the hyper-parameters used for training our models. We fix the gradient norm to be 1 and set the batch size as 1024 in all experiments based on [ 75, 34 ]. Structured pruning can be done by pruning attention heads, pruning encoder units, or pruning the embedding layer. KD is quite different from ours.

As a concrete example, consider a language model that is deployed for sentence completion task in text editors.

Ask an adult what they think about kids and AI, and expect to hear a strong opinion. Parents, politicians, experts--everyone has a take on whether young people should use AI, how to moderate their exposure, and how it's changing the ways they think and communicate. Many of these opinions revolve around education. Adults fret that kids will turn ChatGPT into a research bot, paper writer, or math problem solver. Teachers, specifically, struggle to know how to deal with policing student use, and how to include it in their own pedagogy.

Hon Hai Precision Industry Co. will operate a U.S. factory owned by SoftBank Group Corp., setting up what's in the running to be the first manufacturing site in the Japanese company's 500 billion Stargate venture with OpenAI and Oracle Corp. SoftBank is acquiring Hon Hai's electric-vehicle plant in Ohio but the Taiwanese company will continue to run the complex after turning it into an AI server production plant, Hon Hai Chairman Young Liu said, confirming a report. SoftBank will supply manufacturing gear to the factory, and a joint venture between the two companies will make AI data center-related equipment, Liu said. SoftBank is scouting a number of potential data center sites to serve as a flagship for Stargate, weighing their access to water, power and telecom networks. Hon Hai's participation represents a boon for SoftBank founder Masayoshi Son's ambition to be at the center of surging investment in artificial intelligence hardware. Hon Hai -- known also as Foxconn -- assembles Apple iPhones and Nvidia servers.

Multi-head attention empowers the recent success of transformers, the state-of-the-art models that have achieved remarkable success in sequence modeling and beyond.

The manual design of scenarios for Air Traffic Control (ATC) training is a demanding and time-consuming bottleneck that limits the diversity of simulations available to controllers. To address this, we introduce a novel, end-to-end approach, $\texttt{AirTrafficGen}$, that leverages large language models (LLMs) to automate and control the generation of complex ATC scenarios. Our method uses a purpose-built, graph-based representation to encode sector topology (including airspace geometry, routes, and fixes) into a format LLMs can process. Through rigorous benchmarking, we show that state-of-the-art models like Gemini 2.5 Pro, OpenAI o3, GPT-oss-120b and GPT-5 can generate high-traffic scenarios while maintaining operational realism. Our engineered prompting enables fine-grained control over interaction presence, type, and location. Initial findings suggest these models are also capable of iterative refinement, correcting flawed scenarios based on simple textual feedback. This approach provides a scalable alternative to manual scenario design, addressing the need for a greater volume and variety of ATC training and validation simulations. More broadly, this work showcases the potential of LLMs for complex planning in safety-critical domains.