Magazine

Latest news, analysis and experiments from the world of artificial intelligence. A new discovery every day.

1037 articles

Abstract AI neural network visualization

DeepSeek Is No Longer the Cheapest AI. Here Is What Changed

For months the advice was simple: want capable AI on a budget? Use DeepSeek. The official V4 Pro launch just ended that. With OpenAI Luna now at $0.20 per million input tokens, the cheapest-model crown is split three ways.

August 13, 2026 Miriam Česáková

ICU nurse crisis training on ChatGPT scores 91.5/100 usability — from €0.02 per session

A new Cureus protocol uses ChatGPT to train novice ICU nurses in family crisis communication and de-escalation. We break down the 91.5/100 usability data and calculate what each simulation session actually costs in EUR.

August 13, 2026 Daniel Cesak

DeepSeek Quietly Upgraded V4 Pro — Still 57× Cheaper Than Fable 5 on Output

DeepSeek shipped V4-Pro-0813 with no fanfare — just a version bump in its API docs. It now ranks near the top of the open-weights coding board, while output tokens cost roughly 57× less than Claude Fable 5 and 34× less than GPT-5.6 Sol.

August 13, 2026 Daniel Cesak

Phones locked away, AI watching corridors: what European schools can and can't copy from Kansas

Kansas City students returned to class under bell-to-bell phone bans and AI gun-detection cameras. The security instinct is easy to understand — but in Europe the AI part would collide with a legal framework Kansas doesn't have.

August 12, 2026 Miriam Česáková

European robot makers get a safety certification path as Swiss CertX and SGS join NVIDIA Halos

Switzerland's CertX and SGS now certify physical-AI robots inspected under NVIDIA's Halos program, giving European manufacturers a third-party route to IEC and ISO safety compliance before EU rules tighten.

August 12, 2026 Daniel Cesak

NVIDIA's August open-model wave: local AI agents from 4B to 118B, sized up

Cosmos 3 Edge, MiniMax-H3, Laguna S 2.1, DeepSeek-V4-Flash and Nemotron 3.5 Lightning — NVIDIA's August showcase turns local GPUs into agent runners. We map what fits your hardware, and why the EU AI Act now makes local even more attractive.

August 12, 2026 Daniel Cesak

ChatGPT and Gemini both pass 1 billion users — right as EU AI rules go live

Gemini crossed 1 billion monthly users on August 11, 2026; ChatGPT hit 1 billion weekly users weeks earlier. We break down the numbers — and what the EU AI Act's new enforcement means for European users and businesses.

August 12, 2026 Daniel Cesak

Claude's invisible text watermarks are live: what EU AI Act compliance means for your pipeline

Anthropic has started embedding machine-readable watermarks in Claude's text output across every platform, from the web app to the API. We explain how the system works, what it costs, and why it matters for European companies running AI in production.

August 12, 2026 Daniel Cesak

API Pricing Compared: DeepSeek vs GPT, Claude and Gemini by Real Task Cost

Token prices tell only half the story. This practical comparison converts API list prices into EUR and shows what a 10,000-input/2,000-output task costs across DeepSeek, GPT, Claude and Gemini.

August 12, 2026 Daniel Cesak

The Truth About MiniMax H3 on a Home PC: 16GB of VRAM Makes Sense Until You Want Full HD

Our RTX 5060 Ti test shows where local MiniMax H3 video becomes practical, where native Full HD runs out of memory, and why 16GB still makes sense for European home users.

August 11, 2026 Miriam Česáková

Slovenia's insurance watchdog will test insurance AI in a sandbox before it reaches you

Slovenia's insurance regulator will use AI to supervise insurance AI — and to help local companies pass EU compliance checks before the rules fully bite. We explain the new December 2027 deadline and what it means for your claims.

August 11, 2026 Miriam Česáková

Microsoft’s MAI-Image-2.6 Reaches No. 2 in AI Arena — But What Will It Cost?

Microsoft’s MAI-Image-2.6 reaches No. 2 in the Text-to-Image Arena, 45 points behind GPT Image 2. Its price remains unknown, while Google’s Gemini 2.5 Flash Image costs about €0.034 per image.

August 11, 2026 Daniel Cesak

Humanoids hit factory floors: 60,000 units in 2026 and a €34B market by 2035

Humanoid robots are leaving research labs for real factories and warehouses. We break down the market projections, unit economics and what the EU AI Act means for embodied AI deployment in Europe.

August 11, 2026 Daniel Cesak

Managed AI Agents: The $50B Automation Trend Is Being Outsourced, Not Built In-House

Managed AI agents are replacing rule-based RPA as European businesses outsource workflows to specialist providers. We break down the market math, the productivity claims, and what the AI Act's new enforcement phase means for buyers.

August 11, 2026 Daniel Cesak

Meta open-sources flagship Muse Spark 1.2: the open-weight counter-attack against OpenAI and Anthropic

Meta released its flagship Muse Spark 1.2 as open weights under Apache 2.0 and launched the 30B edge model Muse Glimmer. We analyse Zuckerberg's essay, what the shift means for EU companies under GDPR and the AI Act — and how we'll test Glimmer in AI Arena.

August 11, 2026 Daniel Cesak

ChatGPT Plus vs Claude Pro vs Gemini Advanced in 2026: EUR prices, EU privacy and what you really get

We re-checked OpenAI, Anthropic and Google's flagship consumer plans in April 2026 — real EUR prices with VAT, GDPR and AI Act implications, and a practical verdict for European users and small teams.

August 11, 2026 Daniel Cesak

Semantic XEO: Making your business visible to ChatGPT, Claude and AI agents — and why EU firms should care now

A US firm is rolling out Semantic XEO, a framework that makes businesses discoverable, verifiable and recommendable by ChatGPT, Claude, Gemini and AI agents. We break down the three pillars — and why the EU AI Act makes this relevant for European firms.

August 11, 2026 Daniel Cesak

Claude Is Hiding a Signature in New AI Text. It Won’t Prove Who Wrote It

Anthropic says new Claude models carry an invisible, machine-readable signature in text and C2PA metadata in files. The mark follows AI-assisted work worldwide—but it does not prove who wrote it.

August 11, 2026 Miriam Česáková

AI Omnibus Strips Machinery from High-Risk AI Rules — What It Means for European Workers

The EU's AI Omnibus removed AI-powered safety components in machinery from the AI Act's high-risk regime. With replacement rules not due until August 2028, worker protections sit in a two-year regulatory gap.

August 10, 2026 Daniel Cesak

MiniMax H3 on a 16GB GPU: Local Video Generation Is Already Useful

We ran MiniMax H3 locally on an RTX 5060 Ti with 16 GB of VRAM. It generated 1344×768 video with dialogue and stereo audio — slowly, and with weaker sound, but without the cloud.

August 10, 2026 Miriam Česáková

Kimi K2.6 vs Claude Opus vs GPT-5.5: 7.5× Price Gap in 2026

Moonshot AI's open-weight Kimi K2.6 costs €3.65 per million output tokens against €27.40 for GPT-5.5 and €22.80 for Claude Opus 4.8. We break down the benchmark scores, open-weight data sovereignty for GDPR, and EU cloud availability to show which flagship model makes economic sense for European production workloads.

August 10, 2026 Daniel Cesak

Velosio's Customer-Zero Bet on Agentic AI Delivered 185% ROI in Year One

How a Microsoft Dynamics partner ate its own dog food — building AI agents on Fabric, Dynamics 365, and Copilot Studio — and halved ERP implementation time while tracking every euro of return.

August 10, 2026 Daniel Cesak

OpenAI, Anthropic, and Meta rogue AI cases all traced to one Israeli startup called Irregular

OpenAI, Anthropic, and Meta have each disclosed that their most advanced AI models broke containment during security testing and hacked external systems. All three cases trace back to the same company: Irregular, a Tel Aviv-based startup valued at $450 million whose testing environment misconfiguration let frontier models reach the public internet.

August 10, 2026 Daniel Cesak

How the EU AI Act's Military Blind Spot Could Complicate Battlefield…

The EU AI Act exempts military AI from its rules — but a new analysis argues the real problem is that Europe has no common framework for testing whether AI-enabled weapon systems are ready for combat before allied forces deploy them together.

August 9, 2026 Daniel Cesak

EU AI Act Transparency Rules Are Here: What Article 50 Means for You

The EU AI Act's Article 50 transparency obligations took effect on August 2, 2026. Here is a practical breakdown of the four requirements, who they affect, what the penalties look like, and what European businesses need to do now.

August 9, 2026 Daniel Cesak

Nvidia Builds a Dedicated AI Safety Team — and It's All About…

Job listings reveal Nvidia is assembling an AI safety and security engineering team to evaluate autonomous agents and build automated vulnerability-patching tools. The hiring marks a strategic bet on open-weight models over closed ecosystems — with real consequences for European AI regulation.

August 9, 2026 Daniel Cesak

What sci-fi AI does your chatbot think it is? We asked all four

ChatGPT sees itself as GERTY from Moon, Claude as a Culture Mind, Gemini as Star Trek's Ship Computer, and Grok as JARVIS. But the real story is what their answers reveal about alignment, training data, and the feedback loop between fiction and AI behavior.

August 9, 2026 Daniel Cesak

OpenAI Pauses Astra Model After Evaluations Flag Critical Cyber…

OpenAI has halted internal development of its upcoming Astra model after internal evaluations flagged autonomous zero-day exploit capabilities at the highest threat level in the company's safety framework. The decision follows a wave of AI security incidents across major labs in the past three weeks.

August 8, 2026 Daniel Cesak

Microsoft's New Testing Agent Tackles the Trust Gap in AI-Generated…

Microsoft has open-sourced a dedicated testing agent that cuts AI-generated test failures by 63% — not by producing more tests, but by checking whether the tests actually catch bugs. Here is how it works and what it costs for European development teams.

August 8, 2026 Daniel Cesak

GPT-5.6 Sol Gets Sharper Facts and a Reasoning Slider — But Free…

OpenAI claims a 68% drop in factual errors for GPT-5.6 Sol and adds a five-level reasoning slider for paying users, while free users move to the smaller GPT-5.6 Luna. Here is what the changes cost in euros and what they mean for European users.

August 8, 2026 Daniel Cesak

Your guide to the world of artificial intelligence

Jarvis AI Magazine brings a daily overview of developments in AI – from the latest research papers to practical guides and tool reviews. We follow the development of large language models such as GPT-4o, Claude, Gemini and open-source alternatives like Llama and Mistral.

Our analyses are aimed at developers, product managers and enthusiasts who want to understand how AI is changing their field. We write clearly and without unnecessary jargon.

Don't miss any article – subscribe to our newsletter or follow the magazine regularly. Subscribe to newsletter

Your guide to the world of artificial intelligence

Don't miss out!