The magazine of the European tech scene

Independent since 2024

EuropaNow

Section

AI

AI coverage on Europa Now: models, rules and what European teams actually do with both.

The next thing your endpoint manager manages is an AI agent

AI

Endpoint managers face new challenge: governing AI agents on corporate devices

As AI agents proliferate across enterprise infrastructure, traditional device management tools must evolve from tracking software to controlling what autonomous systems are permitted to do.
Booking.com says it built AI tools it wishes it had bought off the shelf

AI

Booking.com and Diageo on AI regrets: build less, buy more, fix processes first

At HumanX in Amsterdam, two enterprise leaders revealed why they built AI tools they now wish they'd purchased—and what they learned about deploying AI at scale.
Google will launch its first AI chips into orbit next week

AI

Google's AI Chips Head to Orbit in First Suncatcher Test Flight

A prototype satellite equipped with Tensor Processing Units will launch next week aboard SpaceX's Transporter-18 to measure how Google's custom chips withstand the rigours of space.
Meta unveils $1,299 VR glasses and turns its AI glasses into hearing aids

AI

Meta's spring 2027 VR glasses and AI-powered hearing aids reshape the glasses market

At Meta Connect, Mark Zuckerberg unveiled lightweight VR glasses priced at $1,299.99, audio-only Ray-Bans, hearing aid functionality for its glasses, and Muse Charm, a keychain device for its AI assistant.
Google DeepMind’s Ayoub says companies can fence AI in, like banks did

AI

DeepMind's Ayoub: Companies can contain AI with guardrails, not full control

Google DeepMind's vice president of AI technical strategy argues that while enterprises cannot govern AI systems completely, they can establish bounded operational frameworks around them, drawing parallels to how banks managed algorithmic trading decades ago.
OpenAI’s Colin Jarvis says enterprise AI is stuck on deployment, not models

AI

Enterprise AI's Real Bottleneck: Getting Models Out the Door, Not Building Better Ones

OpenAI's deployment engineers say most companies struggling with AI aren't waiting for smarter models—they're stuck figuring out how to actually use the ones they have.
Zendesk and Sierra say AI agents should be paid per result, not per seat

AI

Zendesk and Sierra declare per-seat pricing dead, bet on outcome-based AI agent models

At HumanX in Amsterdam, two customer service AI leaders argued that software licensing is shifting from per-seat to per-result pricing, though they diverged on which companies will thrive in the transition.
US rejects global AI governance at UN Security Council

AI

US blocks unified AI rules at UN, pushes national sovereignty and American tech exports

The White House rejected global governance frameworks for advanced AI at the UN Security Council, arguing instead that each nation should set its own policies while adopting American technology.