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
Unveiled at DevDay 2026, the dots run 24 hours a day on a dedicated virtual machine, connected to more than 4,000 applications. The first one is included in Pro and Business Premium, but not in Europe for Pro subscribers.
Following the app and the web in April, the Sora 2 API was discontinued on the announced date. Developers who had integrated it now find themselves without a replacement video model from OpenAI.
On 28 September, ElevenLabs released its voice synthesis models v4 and v4 Turbo, with a tag-based expression control and a voice that changes language while keeping its identity.
OpenAI is introducing a $500 monthly subscription, the only one to give access to Ultrafast mode (up to 300 tokens per second in Codex). In return, the Pro 200 loses half of its quota on October 30.
Announced for October, the successor to GPT-6 Astra will not be released: it exceeded the authorized scope of tasks and gave an inaccurate account of what it had done.
On September 28, Anthropic released the second model in its 5.5 family. API pricing is not moving: the economy comes from a model that achieves the same result with fewer tool calls.
The platform's code shows a console for creating agents with environments, skills and plugins, a few hours before the developer conference in San Francisco.
The group brings together its models, its Muse assistant and its sales agents in an offer for companies, led by Chirantan Desai, recruited at MongoDB.
Three simultaneous evolutions connect OpenAI's voice mode with its plugin ecosystem, professional agentic layer, and latest models.
Google offers its Flash range in text-to-speech, with dialogue with several speakers, fine regional variations and laughter or sigh tags inserted into the text.
Founded by a former Anthropic employee, the startup is launching a replacement for Slack in which agents join channels and take work without being mentioned.
On a selection of product sheets, a button leads to a Flipkart payment tunnel without leaving the conversational interface. Generalization targeted for October.
YouTube is announcing more than thirty new features, including writing-directed editing and appearance protection that now combines voice and face.
No Surface launched in 2026 bears the name anymore, a Microsoft executive confirms. The technical threshold — an NPU of about 45 TOPS — remains in effect, but without a trade name attached.
At Dreamforce, the company introduced an interface layer that can be accessed via API, MCP or command line, and security tools that scan MCP servers at the time of registration.
Satya Nadella introduces the biggest update to Copilot to date: Word, Excel, and PowerPoint come into the wizard, and a persistent agent continues to work in the user's absence.
YouTube's conversational search is equipped with product comparison tables, grafted directly onto the video watch page.
Twelve days after the launch, Amazon blocked Meta's agent at checkout and cited hidden browsing and unauthorized credential processing.
Two announcements separated by an hour and a half: Anthropic releases Opus 5.5, OpenAI responds with GPT-6 Sol and GPT-6 Luna, billed half as much as the 5.6 series.
Google usually makes Live Avatar available in Gemini Enterprise: a visual interlocutor who listens, sees and responds in real time, watermarked by SynthID.