OpenAI temporarily pauses access to its video model Sora

Here are the top trending news from the world of technology and AI. News that every tech enthusiast should keep a tab on.

1)

OpenAI temporarily pauses access to its video model Sora

In an interesting turn of event, OpenAI has temporarily suspended the access to its text-to-video model Sora. This move comes following a massive outcry by a group of artists, who apparently had leaked the Sora on the Hugging Face platform. The group of artists, referred to as ‘Sora PR Puppets’ claimed that they were initially lured to join as a creative partner for developing Sora. However, they later felt betrayed after realizing that their inputs were allegedly used for marketing PR campaign. Sora remained accessible on Hugging platform for several hours before OpenAI shut down the access. OpenAI later issued a statement saying that it is temporarily shutting down the access to Sora while investigating the situation.

2)  

Google integrates Spotify to its Gemini AI Assistant

Google has connected its AI chatbot assistant Gemini with music app Spotify, allowing users to music playback through voice command as well as keyword research. This feature has already started rolling out in the US market. Users can ask Gemini to play music by specifying the song title, artist, album or playlist. For instance, you can ask Gemini to play ‘My heart will go on’ by Celine Dion and it play the song for you. Additionally, you can also search for tracks using keywords, lyrics or genres. However, it must be noted that despite this integration Gemini does not support creating playlists or starting radio stations on Spotify. This integration only allows for basic playback and search functions.

3)

Claude AI can mimic your unique writing style

Anthropic has introduced a new feature for its Claude AI assistant that allows users to customize the writing style of the AI to better match their own communication preferences. The customization feature aims to give users greater control over how Claude communicates or responds, allowing for responses that align closely with individual needs and contexts. Anthropic provides three preset styles, including Formal for clear and polished text, Concise for shorter and more direct responses and Explanatory for educational replies. However, if these preset styles does not satisfy your requirements, Claude can create custom styles that are trained to imitate other writing styles.

4)     

TikTok to restrict teen access to beauty filters due to mental health worries

TikTok has announced plans to block teenagers in Europe from using beauty filters that alter their facial features, responding to growing concerns about the mental health impacts of such tools. The restrictions will specifically target filters that allow users to change their appearance significantly, such as enlarging eyes, plumping lips, or smoothing skin tones. Filters designed to create humorous effects, like adding dog ears or bunny noses, will remain available. Alongside the filter ban, TikTok is tightening its age verification processes. The platform aims to prevent users under the age of 13 from accessing its platform.

5)

In a challenge to OpenAI’s o1, Alibaba launches Marco-o1 with advanced reasoning capabilities

Alibaba’s researchers have just launched a new advanced reasoning LLM model called Marco-o1, which aims to resolve complex tasks and applications. Following this launch, Alibaba has thrown an open challenge to OpenAI’s recently launched advanced reasoning LLM models o1 and o1 mini. As per Alibaba’s claims, Marco-o1 beats OpenAI’s o1 and o1 mini on several evaluation tests including AIME and MATHS test. Marco-o1 is available to download under permissive license. It contains 32.5 billion parameters and can consider prompts up to 32,000 words in length.