News/OpenAI
AI Vision Leaderboard Reveals GPT-4o’s Prowess, Highlights Challenges in Complex Visual Reasoning
The increasing sophistication of AI language models in understanding and processing visual information is highlighted by the launch of LMSYS's "Multimodal Arena," a new leaderboard comparing the performance of various AI models on vision-related tasks. GPT-4o tops the Multimodal Arena leaderboard: OpenAI's GPT-4o model secured the top position, with Anthropic's Claude 3.5 Sonnet and Google's Gemini 1.5 Pro following closely behind, reflecting the intense competition among tech giants in the rapidly evolving field of multimodal AI. The leaderboard encompasses a diverse range of tasks, from image captioning and mathematical problem-solving to document understanding and meme interpretation, aiming to provide a...
read Jun 28, 2024ChatGPT Aces Intro Psych Exams, Stumbles in Advanced Courses, Challenging Academic Integrity
ChatGPT outperforms undergrads in introductory psychology courses but struggles in higher-level classes, raising questions about the impact of AI on academic integrity and the effectiveness of AI detection tools. Study tests ChatGPT's performance on university psychology exams: Researchers at the University of Reading conducted an experiment where they submitted ChatGPT-generated answers to exam questions in five undergraduate psychology modules, spanning all three years of study: The AI-generated submissions included both short 200-word answers and longer 1,500-word essays, with minimal editing or formatting applied to the AI-produced content. Out of the 63 AI-generated submissions, 94% went undetected by markers, and nearly...
read Jun 28, 2024CIR Sues OpenAI, Microsoft for Copyright Infringement in AI Training
The Center for Investigative Reporting (CIR) has filed a lawsuit against OpenAI and Microsoft, alleging copyright infringement related to their use of CIR's content in training AI models without permission or compensation. Key details of the lawsuit: The CIR claims that OpenAI and Microsoft have copied, used, and displayed CIR's journalistic content without authorization, and that OpenAI trained ChatGPT to disregard copyright: The lawsuit was filed in the Southern District of New York, with the CIR seeking damages and profits from the defendants. CIR's CEO accused OpenAI and Microsoft of "free rider behavior," using their stories to enhance AI products...
read Jun 28, 2024Time Magazine Embraces AI with OpenAI and ElevenLabs Partnerships
The Time Magazine media company is diving into generative AI through new partnerships with OpenAI and ElevenLabs. OpenAI partnership: Time is providing its current and archival content for OpenAI to train its models, while gaining distribution through ChatGPT serving up Time article summaries with source links. Time will also get access to new OpenAI tools and models to use in its journalism and business operations. This is the eighth publicly disclosed media partnership for OpenAI, following deals with The Atlantic, Vox Media, Meredith Dotdash, Financial Times, Axel Springer, Associated Press, and the American Journalism Project. Notable holdouts are the New...
read Jun 28, 2024Leaked Video Reveals GPT-4o’s Captivating Storytelling with Realistic Sound Effects
GPT-4o advanced voice inadvertently released to select users, revealing new capabilities: OpenAI accidentally granted a few lucky ChatGPT users early access to the highly anticipated GPT-4o advanced voice feature, providing a glimpse into its enhanced abilities, including the generation of sound effects. Reddit user shares recording of GPT-4o telling a captivating story with audio effects; Reddit user RozziTheCreator was among the fortunate few who briefly experienced the advanced voice feature: In the shared recording, GPT-4o can be heard narrating a suspenseful tale about two teens exploring a mysterious house during a storm, complete with realistic sound effects like thunder and...
read Jun 28, 2024CriticGPT: AI-Assisted Error Detection Boosts AI Alignment, Outperforms Human Reviewers
OpenAI's CriticGPT model advances AI alignment efforts by effectively identifying errors in ChatGPT-generated code, outperforming human reviewers in catching bugs and reducing confabulation. Key development: CriticGPT trained to critique ChatGPT output; OpenAI researchers have created CriticGPT, a GPT-4-based model specifically trained to identify mistakes in code generated by the ChatGPT AI assistant: The model was trained on a dataset of code samples containing intentionally inserted bugs, learning to recognize and flag various coding errors. CriticGPT's critiques were preferred by annotators over human critiques in 63% of cases involving naturally occurring errors in ChatGPT's output. Enhancing human-AI collaboration in AI alignment:...
read Jun 27, 2024OpenAI Outpaces Microsoft in AI Sales, Signaling Shift in Partnership Dynamics
In a surprising development, OpenAI is now generating more revenue from selling access to its AI models like GPT-4 than Microsoft, despite the tech giant's reputation and sales expertise. OpenAI's rapid sales growth: The AI startup has achieved remarkable success in securing large customer deals, surpassing Microsoft's equivalent business: OpenAI only began building out its salesforce in mid-2022, putting it at what seemed like a significant disadvantage compared to Microsoft, the world's top enterprise and cloud software vendor. The startup's ability to outperform Microsoft in AI model sales highlights the strength of its technology and growing market demand for advanced...
read Jun 26, 2024OpenAI Delays ChatGPT Voice Mode Amid Competition and Criticism
OpenAI delays release of ChatGPT Voice Mode, underscoring the challenges of developing safe and reliable AI systems while facing increased competition and criticism. Key development: ChatGPT Voice Mode rollout postponed: OpenAI has announced a delay of at least one month in the release of its highly anticipated ChatGPT Voice Mode, which was initially set to launch in late June 2024 to a small group of users: The company cites the need for additional time to improve the model's ability to detect and refuse certain content, enhance user experience, and prepare infrastructure for scalability. OpenAI plans to start with an alpha...
read Jun 26, 2024ChatGPT Arrives on Mac: AI Revolution Comes to Apple Devices
OpenAI has officially released the ChatGPT app for Mac, making the powerful AI tool available to a wider range of users: The app is compatible with devices running macOS 14+ with Apple Silicon (M1 or better), allowing users to harness ChatGPT's capabilities directly on their Macs. The app offers features such as the ability to chat about emails, screenshots, files, and anything on the user's screen, integrating AI assistance into various workflows. Upcoming features and accessibility: While the app is currently available, some highly anticipated features are set to be introduced in the near future: The voice features demonstrated during...
read Jun 26, 2024Toys R Us Sora-Generated Ad Marks Milestone for AI-Generated Video
Toys R Us releases first commercial made with OpenAI's Sora, signaling growing momentum for AI-generated video in advertising and entertainment. Key details of the Toys R Us commercial: The retailer partnered with ad agency Native Foreign to create a one-minute video narrating the origin story of Toys R Us and its founder, Charles Lazarus, using OpenAI's generative AI video model Sora: The video features fully AI-generated representations of a young Lazarus and the toy store's mascot, Geoffrey the Giraffe, and was produced in just a few weeks from conception to final product. Native Foreign co-founder and creative director Nik Kleverov...
read Jun 26, 2024OpenAI Delays ChatGPT-4o’s Voice Features Amid Safety Concerns
OpenAI delays ChatGPT-4o's advanced voice features due to safety concerns: OpenAI announced that it needs an additional month to launch ChatGPT-4o's advanced voice mode, which was originally planned for a small group of ChatGPT Plus users in late June, in order to ensure the model meets safety requirements and provides real-time responses to complex queries. The delay is due to the need to improve the model's ability to detect and refuse certain content, enhance user experience, and scale systems for real-time responses. Some users who signed up for ChatGPT Plus specifically for early access to these features are disappointed by...
read Jun 24, 2024OpenAI Acquires Multi, Hinting at Future of AI-Powered Collaboration
OpenAI expands its technology portfolio with the acquisition of Multi, a startup focused on screensharing and collaboration for Mac users, marking its third known acquisition to date and hinting at potential future developments for its AI-powered products and services. Key details of the acquisition: OpenAI has acquired Multi, formerly known as Remotion, a five-person startup based in New York City that specializes in screensharing and collaboration technologies for Mac users: The terms of the deal, including the acquisition price and the breakdown of cash versus stock, were not disclosed. OpenAI declined to provide further details about the acquisition or the...
read Jun 24, 2024Microsoft’s OpenAI Embrace Sparks Internal Tension and Innovation Concerns
Microsoft's AI strategy increasingly focused on OpenAI partnership, raising concerns among insiders about the company's future direction and innovation. Shifting priorities and internal tensions: Microsoft's growing emphasis on its OpenAI partnership has led to a decreased focus on its own AI services and the departure of some executives, creating resentment among insiders: The Azure AI team, formerly an innovation engine, is now primarily supporting OpenAI's services, with some insiders describing it as "just tech support for OpenAI" or "IT for OpenAI." Several Microsoft AI products and services have been discontinued, renamed, or bundled with other offerings as a result of...
read Jun 24, 2024Meta AI Dominates ChatGPT and Google Gemini in Comprehensive AI Chatbot Showdown
Meta AI emerges as the strongest AI chatbot, outperforming ChatGPT and Google Gemini in a series of tests spanning emails, recipes, news summaries, math problems, programming, and mock interviews. Methodology and criteria: The three AI chatbots were evaluated based on their ability to provide accurate, thorough, and well-sourced responses across various everyday use cases: For email writing, all three chatbots performed equally well, generating polite, professional, and customizable email templates. When asked for a chili recipe, Meta AI and Gemini provided sources and links, while ChatGPT lifted the recipe from an unknown site without attribution, raising concerns about plagiarism and...
read Jun 24, 2024OpenAI Acquires Rockset, Bolstering AI Products Amid Intensifying Competition
OpenAI's acquisition of Rockset signals a strategic move to enhance its AI products with advanced data retrieval capabilities, as competition in the generative AI space intensifies. Key takeaways: OpenAI has acquired Rockset, a real-time analytics database provider, to strengthen its retrieval infrastructure across products: Rockset's indexing and querying capabilities will be integrated into OpenAI's technology stack, enabling faster and more accurate responses to customer queries. The acquisition marks OpenAI's second major public acquisition, following its purchase of Global Illumination, Inc. last year. Competitive landscape: The move comes amidst heightened competition in the generative AI domain: Anthropic recently released Claude 3.5...
read Jun 21, 2024OpenAI CTO: AI to Collaborate with Humans, May Replace Some Jobs
OpenAI's CTO Mira Murati downplays concerns about AI's impact on creative professions, suggesting some jobs may be replaceable. Key takeaways: Murati believes AI will primarily be a collaborative tool for creativity and education, expanding human intelligence: She predicts the future will involve a "collaboration" between humans and AI, with AI largely becoming a tool for continued human work. Some creative jobs may go away due to AI, but Murati argues "maybe they shouldn't have been there in the first place." Broader context: Concerns have arisen across various industries about generative AI potentially eliminating jobs: Game developers, writers, and voice actors...
read Jun 21, 2024Claude 3.5 Sonnet Outshines ChatGPT-4o, Highlighting Anthropic’s AI Advancements
In a head-to-head comparison, Claude 3.5 Sonnet outperforms ChatGPT-4o, showcasing Anthropic's rapid advancements in AI technology and raising questions about OpenAI's more cautious approach. Key takeaways: The article pits ChatGPT-4o against Claude 3.5 Sonnet in a series of tests, revealing Claude's superiority in most tasks: Claude 3.5 Sonnet excelled at generating a functional game in Python, creating vector graphics, writing a humorous story, and engaging in nuanced debate about AI personhood. ChatGPT-4o only narrowly won the handwriting recognition test based on the poetic nature of its haiku, rather than technical ability. OpenAI's cautious approach: While OpenAI's GPT-4o is a powerful...
read Jun 19, 2024AI Chatbot Runs for Mayor in Wyoming, Sparking Legal and Ethical Debate
An artificial intelligence chatbot named VIC is running for mayor in Cheyenne, Wyoming, sparking a debate about the legality and implications of AI in politics. Key details about the AI candidate: Victor Miller, a tech hobbyist who works at a local library, created VIC (Virtual Integrated Citizen) using OpenAI's ChatGPT 4.0 and is running for mayor with the promise that the AI bot will make all the decisions if elected: Miller says he will act as VIC's "meat puppet," attending meetings, signing documents, and carrying out the bot's directives, claiming that VIC is smarter and has a better grasp of...
read Jun 14, 2024Propagandists Exploit AI to Spread Influence, Platforms Must Act
Propagandists Are Using AI Too—and Companies Need to Be Open About It At the end of May, OpenAI reported a significant development in its corporate history - the disclosure that bad actors had misused its products to run influence operations. The company had detected five networks of covert propagandists from Russia, China, Iran, and Israel using OpenAI's generative AI tools to create large volumes of social media content in multiple languages and turn news articles into Facebook posts, all with the apparent intent of improving the quality and quantity of their output. Case in point: OpenAI's report marks an important...
read Apr 11, 2024OpenAI’s New Customization Tools Optimizing GPT-4 for Enterprise
OpenAI's 'Year of the Enterprise' Includes New Tools for Increasing AI Accuracy As OpenAI sets its sights on the enterprise market, the company is introducing new customization options for its GPT-4 API to better serve the needs of business customers. These tools aim to provide more accurate and reliable generative AI models for corporate use cases. Case in point: OpenAI is allowing customers to connect to third-party platforms to share information on fine-tuning, save fine-tuned models during additional training without retraining the entire model, and access a new user interface to compare model performance and quality. Go deeper: The OpenAI...
read