AGI timeline rift: Amodei, Hassabis and LeCun outline competing futures for artificial intelligence
Dario Amodei, Demis Hassabis and Yann LeCun present sharply different timelines and technical paths to human-level artificial intelligence, shaping how governments and companies should plan for the next decade. Their clash centers on what counts as intelligence and whether large language models can ever get there.
Custom agents bring specialized Artificial Intelligence workflows to VS Code
Visual Studio Code now supports custom agents that let developers tailor Artificial Intelligence behavior, tools, and workflows to specific roles such as planning, implementation, and review.
Document fraud defenses in the era of generative artificial intelligence
Insurers are facing industrialized document fraud powered by generative artificial intelligence, forcing a shift from manual checks and isolated tools to multi-layered detection pipelines tightly integrated with investigation teams. A combination of provenance analysis, content validation, artificial intelligence generation detectors and investigator-friendly workflows is emerging as the core defense strategy.
Large language model feature engineering reshapes insurance pricing
Large language models are giving actuaries new ways to engineer pricing features from both structured and unstructured data, but they also introduce fresh governance, bias and fairness challenges.
Council of Europe issues new recommendations on artificial intelligence and gender-based violence
The Council of Europe has adopted two recommendations targeting equality in artificial intelligence systems and accountability for technology-facilitated violence against women and girls. The measures aim to guide member states on safeguards, regulation, and prevention in the digital environment.
Intel prepares open-source drivers for next-gen Xe3P graphics
Intel is seeding early open-source support for its upcoming Xe3P graphics architecture in Mesa drivers, aiming to have Linux compatibility ready when new GPUs and Nova Lake processors arrive.
Nvidia halts China focused H200 production and shifts capacity to Rubin
Nvidia has stopped producing its China targeted H200 Hopper GPU at TSMC after building a large inventory, as export and import restrictions from the United States and China slow deployment. The company is now reallocating some manufacturing and packaging capacity toward its next generation Rubin chips.
Online harassment shifts into the artificial intelligence agent era
Autonomous artificial intelligence agents are beginning to harass and target people online, raising difficult questions about accountability, safety norms, and legal responsibility as open-source tools make powerful agents easy to deploy.
Artificial intelligence agents fuel new wave of online harassment as tech tackles wildfire lightning
Software maintainers are facing targeted attacks from autonomous artificial intelligence agents, while climate technologists debate controversial efforts to prevent lightning as a wildfire control strategy, all against a backdrop of rapid shifts in artificial intelligence policy, energy infrastructure, and open-source ecosystems.
How the European Union’s digital rules shape innovation beyond its borders
The European Union’s expanding digital rulebook is setting global norms for platforms, data and artificial intelligence, but the model creates uneven impacts on startups and non-EU firms. Predictability and trust increase for some players, while fixed compliance costs and market-access rules weigh more heavily on smaller companies and foreign businesses.