AI🌪️Thursday Highlights: Bing’s New Search, Meta’s Policy Revisions, Kling AI’s Global Debut, & Mistral’s Latest Model

Welcome to Thursday’s AI roundup! Today’s highlights showcase significant shifts and developments in the world of artificial intelligence. From Bing's transformative AI redesign to Meta facing new scrutiny over its handling of AI-generated explicit content, there’s a lot to unpack. Kling AI has also made a splash with its global launch, directly competing with OpenAI’s Sora, while a new Chinese video model raises concerns about censorship. Join us as we dive into these exciting updates and explore what they mean for the future of AI.

1. Bing’s AI overhaul shifts traditional search results aside

Microsoft Bing is testing a new layout that prioritizes AI-generated answers, pushing traditional search results to the side. This layout offers detailed summaries, videos, and charts for queries, while traditional links appear in a narrow column. Though it enhances information delivery, users hope for an opt-out option. Google has a similar AI feature, but Bing's approach is more extensive. Microsoft claims this new design still drives traffic to websites, maintaining a healthy web ecosystem.

Bing’s AI redesign shoves the usual list of search results to the side

AI summaries are taking over another search engine.

The Verge

2. Meta's Oversight Board calls for better policies on AI-generated explicit images

The Oversight Board urges Meta to refine policies on AI-generated explicit images, recommending a shift from "derogatory" to "non-consensual" terminology and moving these rules to the "Sexual Exploitation" section. They also suggest broader terms for manipulated media and removing the non-commercial clause for image bans. These changes come after cases involving AI-generated images of public figures. The Board emphasizes the need for better user reporting mechanisms and awareness, highlighting the cultural implications and secondary victimization faced by victims. They advocate for flexible, user-focused reporting channels to address the spread of non-consensual imagery effectively.

Oversight Board wants Meta to refine its policies around AI-generated explicit images | TechCrunch

The Board wants Meta to change the terminology it uses for labeling explicit, AI-generated images from “derogatory” to “non-consensual.”

TechCrunchIvan Mehta

3. Kling AI Video Generator goes global, challenging OpenAI’s Sora

Kuaishou's Kling AI has launched globally, ending its China-only restriction. Accessible at KlingAI.com, users get 66 free daily credits to create videos. The platform supports both text-to-video and image-plus-text-to-video generation, putting it in competition with OpenAI's Sora. Kling AI offers fast video creation, averaging 5 minutes per video, and supports three resolutions with various camera controls. It handles multilingual prompts and allows fine-tuning for creativity or relevance. While it excels in realistic sci-fi scenarios, it struggles with some details and consistency in simpler tasks. Future updates will enhance features, including HD production and additional camera movements.

Kling AI Video Generator Now Available Globally to Take on OpenAI’s Sora - Decrypt

Here’s a look at a new AI video generator that offers multilingual support and impressive features.

DecryptJose Antonio Lanz

4. New Chinese video model seen censoring politically sensitive content

Mistral has launched its new flagship model, Large 2, aiming to match the performance of OpenAI and Meta’s latest models. With 123 billion parameters, it excels in code generation, mathematics, and reasoning, outperforming Meta’s Llama 3.1 405B in several benchmarks. Large 2 is designed to minimize hallucinations, providing more accurate and reliable responses. Unlike its competitors, it lacks multimodal capabilities but offers robust multilingual support and handles up to 128,000 tokens per prompt. Available on major platforms like Google Vertex AI and Amazon Bedrock, Large 2 showcases Mistral’s growing influence in the AI field.

Mistral’s Large 2 is its answer to Meta and OpenAI’s latest models | TechCrunch

For frontier AI models, when it rains, it pours. Mistral released a fresh new flagship model on Wednesday, Large 2, which it claims to be on par with the

TechCrunchMaxwell Zeff

That’s a wrap on today’s AI highlights! From Bing’s bold new search interface to Kling AI’s global debut and Meta's policy challenges, it’s clear that the AI landscape is rapidly evolving. Stay tuned for more updates and in-depth analysis on how these advancements will shape the tech world. For continuous coverage and the latest insights, don’t forget to follow us on social media and subscribe to our newsletter. Until next time, keep exploring the cutting edge of AI!

WEB3 AI ML