Skip to content

Large Language Model and Artificial Intelligence Updates for France

AI agents embedded in browsers are circumventing standard security controls, according to Darkreading. The gap leaves enterprise environments exposed as agentic deployments accelerate.

In-house analysis

Capability moved faster than the controls built to contain it. Operators who deploy agentic browsers now own the audit gap, not the vendor.

Read our brief

Latest

See more
Industry||Global

Fujitsu and Scaleway bet on CPU inference for European AI

Fujitsu and Scaleway bet on CPU inference for European AI

Fujitsu and Scaleway announced a strategic collaboration to deploy AI inference on CPU infrastructure inside European data centers. The partnership, formalized with an MoU signed November 27th, pairs Fujitsu's Arm-based FUJITSU-MONAKA CPU platform with Scaleway's cloud infrastructure. The goal is a tested, validated alternative to GPU-centric configurations for AI builders and enterprises.

Continue reading

Fujitsu and Scaleway signed an MoU on November 27th to test CPU-based AI inference in European cloud infrastructure. The partnership combines Fujitsu's Arm-based MONAKA CPU with Scaleway's cloud platform to reduce power consumption and supply chain risk.

In-house analysis

The contrast is capability vs. jurisdiction. GPU clusters are powerful but foreign-controlled. A validated CPU inference path on Scaleway gives European buyers a sovereign option they can actually procure.

Read our brief

Models

See more
Models||Mistral AI

Mistral Moves Coding Agents to the Cloud with Medium 3.5

Mistral Moves Coding Agents to the Cloud with Medium 3.5

Mistral AI has shifted coding agents off the developer's laptop and into the cloud. Mistral Medium 3.5, a 128B dense model released under a modified MIT license, now powers remote agents in Mistral Vibe and Le Chat. Agents run in parallel, asynchronously, and notify the user when complete. Pricing is $1.50 per million input tokens and $7.50 per million output tokens.

Continue reading

Mistral launched remote coding agents in Vibe and Le Chat, powered by Mistral Medium 3.5, a 128B open-weight model. Pricing starts at $1.50 per million input tokens, available now on Pro, Team, and Enterprise plans.

In-house analysis

Capability without control does not sell to enterprise. Mistral is pricing for volume and designing for compliance, which is exactly where European operators will decide whether to buy or build.

Read our brief

Policy

See more
Policy||Nokia Newsroom

Seven European Tech CEOs Demand Unified AI Competitiveness Push

Seven European Tech CEOs Demand Unified AI Competitiveness Push

Seven CEOs spanning Airbus, ASML, Ericsson, Mistral AI, Nokia, SAP, and Siemens published a joint op-ed in early May declaring Europe's technology competitiveness a crisis of its own making. The statement followed a direct meeting with European Commission President Ursula von der Leyen in Brussels. Nokia CEO Justin Hotard initiated the public call alongside the group.

Continue reading

The CEOs of seven major European tech companies, including Mistral AI and ASML, signed a joint op-ed calling Europe's competitiveness decline a self-made crisis. The statement follows a meeting with European Commission President Ursula von der Leyen in Brussels.

In-house analysis

Seven companies can issue a statement. The question is whether Brussels buys coordination or just absorbs the pressure. For France, Mistral AI at the table is the signal worth tracking.

Read our brief

Industry

See more

The European Commission selected Scaleway, Iliad's cloud unit, to build a sovereign public cloud and AI platform for EU institutions. The contract places a French operator at the center of European digital infrastructure.

In-house analysis

Scaleway owns the plumbing; US hyperscalers are locked out. The signal is not the contract announcement. The signal is where EU public-sector distribution just moved.

Read our brief

The brief, by email

The AI moves for France, filtered.

Approved stories, in-house analysis, and the operating consequence. One concise email. No generic roundup.

You subscribe to The LLM Brief, the network behind FrenchLLM.