Hub Artificial Intelligence

Artificial intelligence is either a field of research, the most important technical innovation of mankind, its downfall – or simply a pipe dream from Silicon Valley. Can machines really be intelligent? What is intelligence anyway? What opportunities does AI technology offer, and what are the risks? From neural networks to the science fiction vision of super AI, from deepfakes to AI surveillance: THE DECODER delivers the latest AI news and information on all facets of artificial intelligence.

Artificial Intelligence

Jan 8, 2023Feb 23, 2023

Algorithmic prompting or how to teach math to a large language model

Artificial Intelligence

Nov 19, 2024Nov 19, 2024

Study reveals AI models have hidden capabilities they can't access through normal prompts

Artificial Intelligence

Oct 12, 2024Oct 12, 2024

Apple AI researchers question OpenAI's claims about o1's reasoning capabilities

Artificial Intelligence

Mar 28, 2025Mar 28, 2025

OpenAI outlines new image generation rules for ChatGPT

Artificial Intelligence

Mar 28, 2025Mar 28, 2025

Short

OpenAI has updated GPT-4o to better handle detailed instructions, especially when processing multi-task prompts. The company says the model's programming capabilities and complex problem-solving abilities have also been optimized. The updated version reportedly shows improved intuition, greater "freedom" and creativity, while using fewer emojis in responses. The enhanced language model is now accessible through the API as "chatgpt-4o-latest" and in ChatGPT for subscribers. Free users can expect access to the update in the coming weeks. Artificial Analysis suggests the update places GPT-4o ahead of Claude Sonnet 3.7 in terms of performance.

OpenAI

Artificial Intelligence

Mar 28, 2025Mar 28, 2025

Anthropic's "AI microscope" reveals how Claude plans ahead when generating poetry

Artificial Intelligence

Mar 27, 2025Mar 27, 2025

Update

OpenAI blocks artist names in image prompts but allows studio styles

Artificial Intelligence

Mar 27, 2025Mar 27, 2025

Amazon's latest AI feature turns your hobbies into hyper-specific shopping suggestions

Artificial Intelligence

Mar 27, 2025Mar 27, 2025

Short

Anthropic and Databricks have entered a five-year partnership worth $100 million to jointly sell AI tools to businesses. The collaboration focuses on AI agents designed to handle tasks for humans. According to Databricks CEO Ali Ghodsi, Anthropic's Claude models will be directly available on the Databricks platform and will allow customers to develop their own AI agents using their company data. Both companies face pressure to generate revenue that justifies their substantial valuations - Databricks is valued at $62 billion while Anthropic is valued at $61.5 billion. Databricks said, it is working on improving the accuracy of AI agents to above 95 percent, which would make them truly practical for everyday business use.

WSJ

Artificial Intelligence

Mar 27, 2025Mar 27, 2025

Nvidia G-Assist uses 8 billion parameter model and runs without cloud connection

Artificial Intelligence

Mar 27, 2025Mar 27, 2025

Short

OpenAI CEO Sam Altman announced on X that the company will integrate Anthropic's Model Context Protocol (MCP) into its product lineup. MCP is an open-source standard that helps AI models deliver more relevant and accurate responses by accessing data from various business tools and software applications. The integration is already available in OpenAI's Agents SDK, with support coming soon to the ChatGPT desktop app and Responses API. Since Anthropic released MCP as open source, multiple companies have adopted the standard for their platforms. Anthropic's head of product Mike Krieger welcomed OpenAI's decision to implement MCP, which has become a widely adopted standard with numerous integrations.

Altman via X