AI/ML News

Stay updated with the latest news and articles on artificial intelligence and machine learning

AI Race: power shifts in the model wars

April 2026 turned out to be one of the most explosive months in AI history. OpenAI dropped GPT-5.5, Anthropic sparked debate by withholding Claude Mythos, and new releases from Google, DeepSeek, and other Chinese labs pushed reasoning, agentic capabilities, and multimodality to new heights.

Claude Cowork – your new AI employee

Anthropic’s Cowork marks a major shift from chat-based AI to autonomous digital coworkers that can plan and execute real work directly on your computer. By giving the controlled access to local files, Cowork becomes a practical collaborator for reports, analysis, and file management.

Voice-activated manufacturing: from words to reality in minutes

The Speech-to-Reality system transforms spoken commands into physical objects using a combination of natural language processing, 3D generative AI, and robotic assembly. The system enables users to request items like chairs, stools, or shelves and have them assembled by a robotic arm in as little as five minutes.

AI’s hallucination problem is getting worse

The most advanced AI models from tech giants like OpenAI and DeepSeek are generating false information at unprecedented rates – and no one knows exactly why. Due to this surge in AI “hallucinations”, the reliability of AI across critical fields is being called into question.

Phi-4 – small models, big results

Microsoft’s Phi-4 family is a new generation of compact language models built for complex tasks like math, coding, and planning – often outperforming larger systems. Trained with advanced techniques and curated data, they offer strong reasoning while staying efficient for low-latency use.

GPT-4.5 – a leap forward in AI capabilities

GPT-4.5, OpenAI's most advanced AI yet, features improved natural language understanding, enhanced emotional intelligence, and more intuitive conversations. It excels in writing, brainstorming, and problem-solving while minimizing AI hallucinations for more reliable results.

Microsoft launched the Phi-4 model with fully open weights

Microsoft has launched the Phi-4 model with open weights under the MIT license, offering researchers and developers unprecedented flexibility. With 14 billion parameters, Phi-4 outperforms its counterparts in solving mathematical problems and multitasking, ensuring efficient work with limited resources.

Alibaba vs. OpenAI: Can a new model outperform ChatGPT?

Alibaba's new AI model, QwQ-32B-Preview, challenges ChatGPT with its impressive math and logic skills, outperforming competitors on key benchmarks. Released under an open license, it offers advanced reasoning capabilities but still struggles with tasks requiring strong common-sense understanding.

AI can control computer just like a human

Anthropic has introduced Claude 3.5 Sonnet, a new AI model capable of controlling a computer similarly to a human. The model uses screenshots of the desktop to navigate applications and perform tasks such as clicking, typing, and gathering information.

Stable Diffusion 3.5 opens new doors in digital art

Stable Diffusion 3.5, the latest release from Stability AI, introduces three powerful model variants that deliver enhanced image quality, speed, and accessibility for consumer hardware. The models are free for non-commercial use, and integrate advanced safety features to prevent misuse.

Movie Gen – the future of AI video generation

Meta has unveiled Movie Gen, an AI-powered tool that creates high-definition videos with synchronized sound from simple text prompts. The model provides advanced video creation and editing features, offering users enhanced control over content generation.

Google releases major updates for Gemini models

With price cuts, increased rate limits, and faster output, new Gemini models by Google make advanced AI more accessible for developers worldwide. They boost speed, reduce costs, and enhance performance across a wide range of text, code, and multimodal tasks.

Will Ideogram 2.0 overtake MidJourney?

The latest text-to-image model from Ideogram AI introduces significant advancements that could challenge the dominance of established players like MidJourney and Leonardo AI. New features are already available, including multiple distinct styles, enhanced realism, and advanced prompting tools.

A new era of multimodal AI with GPT-4o

During the Spring Update event OpenAI’s presented GPT-4о – the unique omnimodel that integrates text, audio and image processing, allowing it to work faster and more efficiently than ever before.

Llama 3: the latest advances in LLM

Llama 3, Meta AI's latest advancement, boasts unmatched language understanding, enhancing its capacity for complex tasks. With expanded vocabulary and advanced safety features, the model ensures improved performance and versatility.

Efficient fact-checking in LLMs like ChatGPT with SAFE

Google’s DeepMind developed a new method for long-form factuality in large language models, – Search-Augmented Factuality Evaluator (SAFE). The AI fact-checking tool has demonstrated impressive accuracy rates, outperforming human fact-checkers.

The rise of Grok-1 – a new game-changing LLM

Elon Musk's xAI Corp introduces Grok-1, a new LLM equipped with 314 billion parameters and a Mixture-of-Experts architecture. Released as open source under the Apache 2.0 license, Grok-1 is set to catalyze advancements in AI research.

Stable Diffusion 3 – next-gen AI image generator

Stability AI presented the latest advancement in image generative AI models – Stable Diffusion 3. Its expanded parameter range and diffusion transformer architecture ensure smooth generation of complex, high-quality images and accurate text-to-visual translation.

Google introduces Gemma – a new open-source model

Drawing inspiration from its predecessor Gemini, Gemma is focused on openness and accessibility, offering versatile models suitable for various devices and frameworks. The model marks a significant step towards democratizing AI while emphasizing its responsible development and transparency.

Google’s Gemini AI is going to surpass ChatGPT

A groundbreaking NLP model Gemini AI is set to surpass existing benchmarks. With its multimodal prowess, scalability across various domains, and integration potential within Google's ecosystem, Gemini AI represents a significant leap in AI technology.

No Language Left Behind

Facebook has released the NLLB project (No Language Left Behind). The main feature of this development is the coverage of more than two hundred languages, including rare languages ​​of African and Australian peoples. In addition, Facebook has applied a new approach to the machine learning model, where the translation is carried out directly from one language to another, without intermediate translation into English.