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
▶
Paige Bailey présente Gemma 4, la famille de modèles ouverts de Google DeepMind sous licence Apache 2, de 2 à 31 milliards de paramètres, avec des démonstrations d'inférence entièrement locale dans le navigateur.
A group of female researchers have proposed ‘Wellbeing-Former’, a model that scores levels of distress, disruption to daily life and mistrust of institutions as expressed on social media, whilst providing a rationale for each score.
A simulation-based inference method, which had proved reliable with synthetic data, failed in the case of WASP-39b. The problem lay not with the physics of the simulator but with an overlooked variable: MIRAGE incorporates this variable and reduces the χ²/N from 301 to 0.06.
Nolla Health has been granted authorisation by the Utah regulator to prescribe topical treatments for mild to moderate acne, at a cost of $4.99 per month, under medical supervision that will be gradually relaxed.
The AI agent answers questions about products, sizes or delivery and remembers the user’s preferences, whilst Buy Direct allows you to shop with a brand without leaving the video feed.
The Munich-based manufacturer of industrial robotic arms has completed a $40 million funding round – mainly comprising shares sold by employees – led by Sequoia and Lightspeed, to accelerate its growth in the United States.
The Internet Watch Foundation assessed 6,310 AI-generated images that fell within the legal definition of child sexual abuse material in the first half of 2026, compared with just over 4,500 for the whole of the previous year.
Led by Andreessen Horowitz, the funding round supports Core, a screenless device costing $3,499 and fitted with an Nvidia Blackwell GPU, designed to continuously run agents that carry out tasks on behalf of its owner.
The New York-based start-up founded by former DeepMind employees claims to match GLM-5.2 and come close to Qwen 3.8-Max in terms of code and agent-based tasks, whilst using three to four times less computational power.
An official from the US Department of Defence confirmed this to the BBC, whilst Claude was still on active duty last week, according to several sources, including those involved in operations against Iran.
▶
Le PDG de Sourcegraph explique pourquoi le flot de code produit par les agents accélère la dégradation des grandes bases de code, et pourquoi le contexte compte plus que la qualité du modèle.
On a self-paced learning platform, requiring users to book study slots in advance increases the frequency from 1.01 to 1.33 active days per week, whilst making the server load predictable.
A study involving 15 participants shows that interactive risk prediction tools are better understood than expected, but that technical jargon and the way risk is framed remain obstacles.
Thousands of simulated light curves show that an autoregressive detection method would identify at least half of the short-period, Earth-sized planets after just three months of observation.
A study applies post-training quantification to two global forecasting AI models, DLWP and FourCastNet, to reduce computational load and energy consumption without compromising short-term forecasts.
Researchers have proposed a flexible capacitive skin covering the entire hand, with the same arrangement of 202 sensors on both humans and robots, to train robots using human demonstrations.
Fewer than 40 per cent of users of programming agents isolate them in a sandbox. A researcher has proposed SideKernel, a local, open-source micro-VM designed first and foremost to be easy to use.
Two former engineers who are shareholders accuse Groq’s board of having sold off the bulk of the company to Nvidia without a shareholder vote and at a bargain price. Groq considers the complaint to be unfounded.
Seven months after its Series A funding round, the Zurich-based start-up has completed a Series B round led by Scale Venture Partners to accelerate its product development and expand its own computing infrastructure.
From the end of October, OpenAI will be trialling an illustrated advertising format in the US – displayed whilst ChatGPT generates an image – and is enhancing its analytics tools for advertisers.