AM-IA brings together news, research papers and applications relating to artificial intelligence published around the world into a single feed — and makes them available in your European language at the click of a button.
Scroll to browse · Click to filter
Anthropic’s fifth office in the Asia-Pacific region, the Singapore office, is accompanied by a Memorandum of Understanding (MOU) with the Ministry of Digital Affairs covering model evaluation and responsible adoption.
Bessemer and Lightspeed are funding this messaging service, which is aimed at frontline staff and has already been adopted by companies such as FedEx, Starbucks and Taco Bell.
Having worked on the Model 3, Apple’s car project and Ford’s electric vehicle strategy, Doug Field is taking on the role of chief operating officer at a start-up that has already raised over a billion dollars.
Rather than perturbing a single forecast, OceanFlow directly learns the distribution of future ocean states around a stable mean trajectory.
By combining YOLO detection, particle filtering and monocular depth estimation, researchers have identified 82 per cent of occlusions and estimated depth to within one centimetre.
By classifying ten equations of state based on noisy data from several detectors, the study shows that the geometry of the network of observatories has a significant impact on the result.
Although almost perfect on its training data set, a neural network-based intrusion detector fails as soon as the composition of the traffic changes even slightly.
Pavlo Dral and his colleagues propose learning how to generalise from one quantum computing method to another, in order to make predictions that go beyond the quality of the training data.
The three models tested all tend to produce average marks, over-rating weaker papers and under-rating the best ones – a common error that the aggregation process does not correct.
▶
Matt Wolfe met Gemini 4 Argon à l'épreuve, de la création de jeux à l'impression 3D, et le compare à Claude et ChatGPT.
Two weeks after a launch restricted to cybersecurity professionals, Google is making its Edge model available to developers, businesses and Pro and Ultra subscribers.
Following on from a presidential decree issued in September, the scientific funding agency is urging its researchers to replace ‘AI’ with ‘SI’, going beyond what the decree requires.
Researchers hide malicious commands, encoded using the Caesar cipher, within third-party skill files: agents then decrypt them and execute them.
Researchers propose a scale from L0 to L5 to characterise the self-improvement of robots and note that no published work actually reaches the recursive stage.
The French advertising group has exceeded market expectations, confirmed its annual targets and noted that clients are keen to embrace AI in marketing but remain cautious about their major transformation projects.
In *Noema*, the auditor Anna August argues that the registration of models, the prompt reporting of incidents and independent audits would be sufficient to make AI transparent.
A longitudinal study links an Alzheimer’s disease progression score to changes in ‘brain age’ estimated using covariance neural networks.
To find out whether a machine learning model is better at predicting turbulence above large telescopes, a point of comparison is needed: one team proposes two such points.
According to the Institute of Student Employers, 92 per cent of employers have suspected or detected fraud when recruiting graduates, with the undeclared use of AI topping the list.
▶
Cinq actualités en cinq minutes : les modèles Muse Wonder de Meta, AlphaProof Nexus validé par les pairs, l'accès de l'AISI britannique rétabli, les puces Maia de Microsoft et le virage d'Epic Games.