China’s largest cloud computing company, Alibaba Cloud, has released a new computer vision model, Qwen2-VL, which the company says improves on its predecessor in visual understanding, including video comprehension and text-to-image processing in languages including English, Japanese, French, Spanish, Chinese and others. The company says it can analyze videos of more than 20 minutes in length and is able to respond appropriately to questions about content. Third-party benchmark tests compare Qwen2-VL favorably to leading competitors and the company is releasing two open-source versions with a larger private model to come. Continue reading Alibaba’s Latest Vision Model Has Advanced Video Capability

An AI-powered coding app called Cursor is building a fanbase, with everyone from hobbyists to engineers subscribing to the service. The platform reportedly has 30,000 paying customers, among them employees at OpenAI, Midjourney and Perplexity. Referred to as “the ChatGPT of coding,” Cursor uses popular models including GPT-4o and Claude 3.5 Sonnet to automate building apps and other coding tasks. Cursor was launched by two-year-old startup Anysphere, which has raised more than $60 million in Series A funding led by Andreessen Horowitz and Thrive Capital. Continue reading New AI Coding App Cursor Gains Following and $60M in Funds

In a move toward increased transparency, San Francisco-based AI startup Anthropic has published the system prompts for three of its most recent large language models: Claude 3 Opus, Claude 3.5 Sonnet and Claude 3 Haiku. The information is now available on the web and in the Claude iOS and Android apps. The prompts are instruction sets that reveal what the models can and cannot do. Anthropic says it will regularly update the information, emphasizing that evolving system prompts do not affect the API. Examples of Claude’s prompts include “Claude cannot open URLs, links, or videos” and, when dealing with images, “avoid identifying or naming any humans.” Continue reading Anthropic Publishes Claude Prompts, Sharing How AI ‘Thinks’

OpenAI announced its newest model, GPT-4o, can now be customized. The company said that the ability to fine-tune the multimodal GPT-4o has been “one of the most requested features from developers.” Customization can move the model toward more specific structure and tone of responses or allow it to follow specific instruction sets geared toward individual use cases. Developers can now implement custom datasets, aiming for better performance at a lower cost. The ChatGPT maker is rolling out the welcome mat by offering 1 million training tokens per day “for free for every organization” through September 23. Continue reading OpenAI Pushes GPT-4o Customization with Free Token Offer

Meta Platforms CEO Mark Zuckerberg and Spotify CEO Daniel Ek have joined forces to express displeasure with the European Union’s regulations on artificial intelligence, claiming they are suppressing innovation. That is the opposite of the stated goals of EU lawmakers in passing the regulations. In a joint statement first published in The Economist and then on the Meta and Spotify websites Friday, the duo took aim at alleged EU obstruction to the development of open source AI, suggesting that Europe’s “fragmented regulatory structure, riddled with inconsistent implementation, is hampering innovation and holding back developers.” Continue reading Meta, Spotify Issue Statement Criticizing EU’s AI Regulations

YouTube is testing an integration with parent company Google’s Gemini AI. Called Brainstorm with Gemini, it invites creators to ideate with video, titles and thumbnails. The limited test makes the feature available to a handful of creators whose feedback will be used in strategizing how and whether to introduce the feature more broadly. In May, YouTube began testing another AI tool, renaming its “Research” tab “Inspiration.” The Inspiration tool provides topics that its algorithm detects a creator’s audience might find interested, supplying an outline and talking points. Brainstorm is similar but supports Google’s AI branding. Continue reading YouTube Invites Content Creators to ‘Brainstorm with Gemini’

Google has unveiled three additions to its Gemma 2 family of compact yet powerful open-source AI models, emphasizing safety and transparency. The company’s Gemma 2 2B is a 2.6 billion parameter update to the lightweight 2B parameter Gemma 2, with built-in improvements in safety and performance. Built on Gemma 2, ShieldGemma is a suite of safety content classifier models that “filter the input and outputs of AI models and keep the user safe.” Interoperability model tool Gemma Scope offers what Google calls “unparalleled insight into our models’ inner workings.” Continue reading Latest Gemma 2 Models Emphasize Security and Performance

OpenAI has released its new Advanced Voice Mode in a limited alpha rollout for select ChatGPT Plus users. The feature, which is being implemented for the ChatGPT mobile app on Android and iOS, aims for more natural dialogue with the AI chatbot. Powered by GPT-4o, which is multimodal, Advanced Voice Mode is said to be able to sense emotional inflections, including excitement, sadness or singing. According to an OpenAI post on X, the company plans to “continue to add more people on a rolling basis” so that everyone using ChatGPT Plus will have access to the new feature in the fall. Continue reading OpenAI Brings Advanced Voice Mode Feature to ChatGPT Plus

Microsoft has begun the release of Bing generative search, making it available for “a small percentage of user queries.” The company says it will solicit user feedback and undertake further testing prior to a broader rollout. Google began dabbling in what it called the Search Generative Experience last summer, then upped the ante by adding a search-optimized version of its Gemini model this spring. The journey was not without controversy, something Microsoft will surely try to avoid. Microsoft says its new AI-driven search functionality “combines the foundation of Bing’s search results with the power of large and small language models (LLMs and SLMs).” Continue reading Microsoft Testing Bing Generative Search for User Feedback

Airtable, a 10-year-old firm focused on customized apps, is launching Cobuilder, which uses AI to turn a concept into a customizable application “in seconds,” without the need for human coding. The debut adds to a rapidly expanding field of no-code platforms that help non-technical types develop software suitable for enterprise use. “Within the next five years, teams will build the vast majority of applications in-house, customizing them to transform their most critical workflows,” predicts Airtable co-founder and CEO Howie Liu. “To get there, knowledge workers who are closest to the work need to be empowered to build.” Continue reading Airtable Enters No-Code Enterprise App Space with Cobuilder

Meta Platforms is bringing its Meta AI assistant to Quest 3 headsets in the U.S. and Canada. As part of the company’s ongoing collaboration with Microsoft, Quest 3’s Horizon OS will also get access to real-time information courtesy of Bing search. The update will also bring Meta AI with Vision — already integrated with the Ray-Ban Meta smart glasses — to Horizon OS, initially just in passthrough mode. “These changes will begin rolling out in the U.S. and Canada in experimental mode next month — and we’ll continue to improve the experience over time,” Meta explains. Continue reading Meta Brings AI Assistant, AI Vision to Some Quest Headsets

Nvidia and French startup Mistral AI are jointly releasing a new language model called Mistral NeMo 12B that brings enterprise AI capabilities to the desktop without the need for major cloud resources. Developers can easily customize and deploy the new LLM for applications supporting chatbots, multilingual tasks, coding and summarization, according to Nvidia. “NeMo 12B offers a large context window of up to 128k tokens,” explains Mistral, adding that “its reasoning, world knowledge, and coding accuracy are state-of-the-art in its size category.” Available under the Apache 2.0 license, it is easy to implement as a drop-in replacement for Mistral 7B. Continue reading Mistral, Nvidia Bring Enterprise AI to Desktop with NeMo 12B

Google has launched the beta version of its Gemini-powered Google Vids productivity app, which lets users create work-related video presentations that embed documents, slides, audio recordings and even additional videos into a timeline. Incorporated into Workspace Labs, Google’s AI preview space, Google says invited participants can use Vids to “build a narrative with high quality templates” or “get to a first draft faster.” Access to Google’s royalty-free stock content library and Vids recording studio means a project can be completed “without ever leaving Workspace,” according to the company. Continue reading Gemini Powering Google Vids Multimedia Presentation Builder

Microsoft is working on a new productivity tool that helps artificial intelligence better understand spreadsheets. Still in the experimental phase, SpreadsheetLLM addresses challenges that are unique to applying AI to spreadsheets, “with their extensive two-dimensional grids, various layouts, and diverse formatting options,” the company explains. Hailed as a significant development in the enterprise space, where spreadsheets are used for everything from data entry to financial modeling and are shared among departments, Microsoft points out that as a research area spreadsheet-optimized AI has generally been overlooked in favor of flashier use-cases. Continue reading Microsoft Targets Enterprise Productivity with Spreadsheet AI

After five months of testing, e-commerce giant Amazon is releasing its AI-powered shopping assistant, Rufus, on mobile throughout the U.S. via the Amazon Shopping app. By tapping the icon, shoppers can access a chat interface and query Rufus on topics such as the best portable speakers or summer dresses under $50. In time for this week’s Prime Day event, the new assistant can also provide status updates on orders. Rufus was trained on the Amazon catalog and other Internet content, so it can provide information on a wide variety of topics and reportedly also answer questions about politics and write short stories. Continue reading Amazon’s AI Shopping Assistant Rufus Is Ready for Prime Day