Jorick van Weelie
Marketing Lead at DataNorth | AI Enthusiast & Tech Storyteller
Articles written by Jorick van Weelie
Alibaba’s Qwen4-Max is open for testing until 25 November, with
Alibaba opened a two-week preview of Qwen4-Max, the top model of its next Qwen generation, on 11 November 2026. Developers can use it through a Token Plan subscription from $3
ChatGPT now lets businesses claim their profile and see how
OpenAI opened Businesses in ChatGPT as a beta on 10 November 2026. In Business Hub, companies with a website can claim the profile ChatGPT shows, correct it and see 28
Gemini now lets you edit the documents, slides and PDFs
On 29 October 2026, Google turned the Gemini app into a file editor. Documents, slide decks, spreadsheets and PDFs that Gemini generates can now be changed inside the chat, by
Gemini 4 Carbon ranks first on Artificial Analysis, but only
Google DeepMind introduced Gemini 4 Carbon on 28 October 2026 as its strongest model. It scores 59 on the Artificial Analysis Intelligence Index, just ahead of Claude Opus 5.5 at
ChatGPT’s dot agents now take tasks by text message, starting
On 28 October 2026 OpenAI opened texting for ChatGPT dots, its always-on agents. US Pro subscribers can now send tasks over iMessage or RCS and get proactive updates back, while
PHP2Shell: AI found holes in every PHP version since 2012,
On 21 October, Wiz disclosed PHP2Shell, two critical PHP flaws it found with Google's Gemini 4 Argon and OpenAI's GPT-5.6 Sol. Attackers can use them to take over web servers
Microsoft lets an EU partner oversee engineer access for European
On 20 October 2026 Microsoft started rolling out Data Guardian for cloud customers in the EU and EFTA. A trusted EU partner, which Microsoft does not name, oversees qualifying remote
Model Hardware Standard (MHS): What it is, How it works,
This guide explains how MHS works, how it compares to MCP and existing standards, what the pilots showed, and what is still planned.
Mistral’s Timestral forecasts on a CPU, but its free licence
Mistral released Timestral on 15 October 2026: four open models, from 5M to 209M parameters, that forecast sales, energy or sensor data without training and run on an ordinary CPU.
ChatGPT Voice gets GPT-Live-1.1, with 13 new voices in seven
OpenAI started rolling out GPT-Live-1.1 in ChatGPT Voice on 15 October 2026. The update brings 13 new voices for French, German, Hindi, Japanese, Korean, Mexican Spanish and English.
AI effort levels explained: when to use low, medium, high
Effort levels decide how long an AI model thinks before it answers, and that choice changes quality, speed and cost at the same time. This guide explains how ChatGPT, Claude,
Claude Haiku 5.5 matches GPT-6 Luna’s price and beats it
Anthropic released Claude Haiku 5.5 on 7 October 2026. It costs $0.10 per million input tokens and $0.50 per million output tokens, the same as GPT-6 Luna, and in Anthropic's
ChatGPT’s textGrain watermark arrives in the EU
OpenAI announced on 5 October 2026 that ChatGPT and Codex will watermark the text they write for EU users, on every plan, to meet the AI Act's marking rule. The
What it costs to keep an AI system running after
Running an AI system costs four things: cloud and model usage, maintenance hours, ownership and coordination, and retraining where a trained model is involved.
Cloudflare releases Clef and Clef-flash
Cloudflare released Clef and Clef-flash on 1 October 2026, two open-weight decision models that return probabilities instead of text. Clef-flash decides in 38.8 milliseconds at the median, both carry a
Microsoft releases MAI-Transcribe-2-Streaming
Microsoft released MAI-Transcribe-2-Streaming on 1 October 2026, its first real-time transcription model, with MAI-Voice-2.1 and MAI-Voice-2.1-Flash. Microsoft reports a 2.5% word error rate and a final transcript 0.13 seconds after
Perplexity releases pplx-embed-v2-context-9b-preview
Perplexity released pplx-embed-v2-context-9b-preview on 30 September 2026, an MIT-licensed embedding model with open weights on Hugging Face.
Liquid AI releases d1
Liquid AI released d1 on 29 September 2026, a decision model that classifies, routes and scores without generating a single output token.
ElevenLabs releases Eleven v4 and Eleven v4 Turbo
ElevenLabs released Eleven v4 and Eleven v4 Turbo on 28 September 2026. Turbo starts speaking after about 150 milliseconds, against 262 for Cartesia Sonic 3.6 and 814 for OpenAI's GPT-4o
MiniMax releases M3.1-Flash-Preview
MiniMax released M3.1-Flash-Preview on 27 September 2026, a coding model with a 1,000,000-token context that runs only inside the company's MiniMax Code tool.
