OpenAI

OpenAI's premium users now have access to GPT-4 Turbo (trained on data up until December 2023), which promises enhanced capabilities in writing, mathematics, logical reasoning, and coding. OpenAI says the model will provide more concise, conversational responses compared to its predecessors. Continue Reading →

YouTube CEO Neal Mohan has a clear message for OpenAI: using the platform's videos to train AI models (like Sora) without permission is a "clear violation" of YouTube's terms of service. Continue Reading →

The game of LLMusical chairs continues. Emad Mostaque, CEO of Stability AI (makers of the text-to-image app Stable Diffusion), has stepped down. He tweeted: "The concentration of power in AI is bad for us all. I decided to step down to fix this at Stability & elsewhere." Continue Reading →

Elon Musk's xAI has officially open-sourced its AI chatbot Grok, which is now available on GitHub. The release includes the base model weights and network architecture of Grok-1. Continue Reading →

OpenAI's Sora, an almost unbelievable text-to-video generator, has been making waves since it was announced. In February, Sora was made available to a select group of "red teamers" for security and stability testing, as well as to a limited number of visual artists, designers, and filmmakers for feedback on its video generation capabilities. Continue Reading →

According to Inflection AI, Inflection-2.5 performs at more than 94% the average performance of GPT-4, despite using only 40% of the training FLOPs (floating point operations per second). Continue Reading →

Anthropic claims that Claude 3, the company's most recent AI release, has achieved "near-human" capabilities in various cognitive tasks. It's a bold claim. Let's put it in perspective. Continue Reading →

OpenAI has accused The New York Times of employing deceptive tactics to generate evidence for a copyright lawsuit against the AI company. In a legal filing in Manhattan federal court, OpenAI alleges that The New York Times used "deceptive prompts" to make ChatGPT reproduce the newspaper's content, which OpenAI argues violates its terms of use and undermines the integrity of the legal process. Continue Reading →

Paris-based startup Mistral has unveiled Mistral Large, a new open source AI model that builds on its initial offering, Mistral 7B. Designed for developers, Mistral Large distinguishes itself with exceptional reasoning skills and proficiency in code and mathematics. The model is multilingual and is fluent English, French, German, Spanish, and Italian. Continue Reading →

Yann LeCun, Meta's Chief AI Scientist, has been all over social media lately. He's looking to reboot the AI research echo chamber, challenging the current approach to AI championed by OpenAI, Google, and other big players. Continue Reading →