IBL News

Category: AI Literacy

  • Runway Issues Its Model ‘Gen-4’, Which Allows to Generate Consistent Characters

    Runway Issues Its Model ‘Gen-4’, Which Allows to Generate Consistent Characters

    IBL News | New York

    Runway released its most advanced video generator model called Gen-4 for paid and enterprise customers last week.

    The AI video tools startup claimed it can generate consistent characters, locations, and objects across scenes, maintain coherent world environments, and regenerate elements from different perspectives and positions within scenes “without the need for fine-tuning or additional training.”

    To craft a scene, users can provide images of subjects and describe the composition of the shot they want to generate.

    On the other hand, Runway AI Inc. announced that it raised $308 million in a new round of funding, which more than doubled the company’s valuation. The deal pushes Runway’s value to just over $3 billion. Private equity firm General Atlantic led the round, which closed late last year. Other investors included Nvidia Corp. and SoftBank Group Corp.’s Vision Fund 2.

    The company has been able to differentiate itself, inking a deal with a major Hollywood studio and earmarking millions of dollars to fund films using AI-generated video.

    Runway says that Gen-4 allows users to generate consistent characters across lighting conditions using a reference image of those characters.

    “Runway Gen-4 [also] represents a significant milestone in the ability of visual generative models to simulate real-world physics,” said the company.

    Like all video-generating models, Gen-4 was trained on many video examples to learn the patterns and generate synthetic footage.

    Runway refused to say where the training data came from, out of fear of sacrificing competitive advantage and also to avoid IP-related lawsuits.

     

  • Anthropic Launches ‘Claude for Education’ Program to Compete with OpenAI

    Anthropic Launches ‘Claude for Education’ Program to Compete with OpenAI

    IBL News | New York

    Anthropic launched a specialized version of Claude tailored for higher education institutions this week to answer OpenAI’s ChatGPT Edu plan.

    The Claude for Education initiative seeks to equip universities with AI-enabled approaches to teaching, learning, and administration.

    With this program, Anthropic is trying to boost its revenue in the university space, where it competes with OpenAI. The company already reportedly brings in $115 million a month.

    Claude for Education includes what Anthropic calls a Learning mode. This mode is based on guiding students’ reasoning process rather than providing answers, helping them develop critical thinking skills. This feature works within Projects and saved conversations, where students can organize their work around specific assignments or topics.

    The solution will be embedded into CanvasLMS and extended with a program called Claude Campus Ambassadors, which will offer API credits for students who build projects.

    Anthropic said it collaborates with Northeastern University, the London School of Economics and Political Science (LSE), and Champlain College.

    The AI start-up summarized its offer in these terms:

    • “Students can draft literature reviews with proper citations, work through calculus problems with step-by-step guidance, and get feedback on thesis statements before final submission.
    • Faculty can create rubrics aligned to specific learning outcomes, provide individualized feedback on student essays efficiently, and generate chemistry equations with varying difficulty levels.
    • Administrative staff can analyze enrollment trends across departments, automate repetitive email responses to common inquiries, and convert dense policy documents into accessible FAQ formats—all from a familiar chat interface with enterprise-grade security and privacy controls.” 
  • OpenAI’s Studio Ghibli Style Generated Images Flood Social Media with Memes

    OpenAI’s Studio Ghibli Style Generated Images Flood Social Media with Memes

    IBL News | New York

    AI-generated memes in the style of Studio Ghibli—the cult-favorite Japanese animation studio behind films such as “My Neighbor Totoro” and “Spirited Away”—are flooding social media, forcing OpenAI to put a rate limit on image generation requests, according to CEO Sam Altman.

    “It’s super fun seeing people love images in ChatGPT, but our GPUs are melting,” he posted on X today.

    The newly improved image generation of GPT-4o, released this week, is resulting in creations that are more realistic than before, and even users can take them in any number of directions.

    Usually, users upload existing images and pictures into ChatGPT and ask the chatbot to recreate them in new styles. OpenAI’s and Google’s latest tools make it easier than ever to re-create the styles of copyrighted works — simply by typing a text prompt.

  • Mistral Releases an Open Source Model that Outperforms Gemma 3 and GPT-4o Mini

    Mistral Releases an Open Source Model that Outperforms Gemma 3 and GPT-4o Mini

    IBL News | New York

    Paris–based Mistral AI unveiled Mistral Small 3.1, a new multimodal open-source model. According to the company, it is “the best model in its weight class ” and “outperforms comparable models like Gemma 3 and GPT-4o Mini. “

    Released under an Apache 2.0 license, Mistral Small 3.1 has an expanded context window of up to 128k tokens and a delivery inference speed of 150 tokens per second.

    Experts say that Mistral Small 3 is competitive with larger models such as Llama 3.3 70B or Qwen 32B and replaces opaque proprietary models like GPT4o-mini.

    Mistral Small 3 can be fine-tuned to specialize in specific domains, creating highly accurate experts. This is particularly useful in fields like legal advice, medical diagnostics, and technical support, where domain-specific knowledge is essential.

    This model sets the stage for increased competition in a market dominated by U.S. tech giants. Mistral’s open-source approach highlights a growing divide in the AI industry between closed, proprietary systems and open, accessible alternatives.

    After raising $1.04 billion, founded in 2023 by former researchers from Google DeepMind and Meta, Mistral AI has rapidly established itself as Europe’s leading AI startup, with a valuation of approximately $6 billion. While impressive for a European startup, this valuation remains a fraction of OpenAI’s reported $80 billion.

    Mistral Small 3 Human Evals

    Mistral Small 3.1 joins the company’s rapidly expanding suite of AI products.

    Earlier this month, the company introduced Mistral OCR, an optical character recognition API that converts PDF documents into AI-ready Markdown files. This addresses a critical need for enterprises seeking to make document repositories accessible to AI systems.

    These specialized tools complement Mistral’s broader portfolio, which includes Mistral Large 2 (their flagship large language model), Pixtral (for multimodal applications), Codestral (for code generation), and “Les Ministraux,” a family of models optimized for edge devices.

  • Google Gemini’s Latest Feature: Canvas, an Interactive Space for Refining Documents and Code

    Google Gemini’s Latest Feature: Canvas, an Interactive Space for Refining Documents and Code

    IBL News | New York

    Google added this week a dedicated workspace to its Gemini chatbot called Canvas, an identical name used by OpenAI for the same feature – and similar to Anthropic’s Artifacts.

    It’s an interactive space where users can refine documents, create and debug code, and share writing and coding projects while using Gemini’s feedback to suggest edits and adjust the tone, length, or formatting.

    Canvas also streamlines the process of transforming coding ideas into working prototypes for web apps, Python scripts, games, simulations, and other interactive apps.

    It can also generate and preview HTML/React code and other web app prototypes to visualize the design, such as a website’s email subscription form.

    This feature works by selecting Canvas in the prompt bar and start creating. Google created a dedicated website.

    Gemini’s Canvas includes the Audio Overview of NotebookLM, which went viral last year. It transforms users’ files into realistic-sounding podcast-style discussions between two AI hosts, with audio summaries of documents, web pages, and other sources.

    Uploading a document via the prompt bar triggers the Audio Overview shortcut. Once a summary is generated, it can be downloaded or shared via the Gemini app on the web or mobile.

    Gemini Canvas