AI Industry News and Updates
Curated by the AI BOOM World team, augmented by AI
Gumloop | AI Automation Framework
Gumloop introduces an AI Automation Framework that allows businesses to automate workflows without requiring AI expertise. The platform offers drag-and-drop tools for tasks like scraping websites and writing to Google Sheets. It is backed by Combinator and emphasizes security with features like SOC 2 & GDPR compliance, encryption, and access control. Scalability features include auto-scaling compute and parallelized execution. Users can connect nodes, pass data through drag-and-click connections, and test workflows in a sandbox environment. Gumloop aims to simplify automation for enterprises while prioritizing data security and scalability.
Source:https://www.gumloop.com/
Visual Electric - AI Image Generator
Visual Electric, an AI image generator, has received various creative prompts for generating images, including fashion photography, floral compositions, underwater scenes, and urban portraits. The prompts range from avant-garde fashion shoots to serene floral arrangements, showcasing the versatility and capabilities of the AI technology.
Source:https://visualelectric.com/
Relume — Websites designed & built faster with AI | AI website builder
Relume offers an AI website builder that helps design and build websites faster. Users can generate sitemaps, turn them into wireframes, and create UI style guides with AI assistance. The platform is used by leading brands globally and allows for smart tagging, easy collaboration, and seamless export to tools like Figma and Webflow. With over 250,000 websites designed using Relume, it provides a library of components for efficient website development.
Source:https://www.relume.io/
Visily - AI-powered UI design software
Visily is an AI-powered UI design software that allows users to easily create hi-fidelity wireframes and prototypes without a learning curve. The software leverages AI to simplify design processes and offers features like Text to Diagram Beta, Screenshot to Design, and Sketch to Design. With over 1,500 prebuilt templates, Visily aims to streamline the design workflow for individuals and teams. Testimonials from professionals highlight the software's user-friendly interface and collaborative capabilities.
Source:https://www.visily.ai/
Clay - Uplevel your data enrichment. Scale personalized outreach.
Clay offers data enrichment and personalized outreach solutions through its platform. Users can import or create targeted lists, access 75+ enrichment tools, and automate manual research workflows with AI agents. The platform also provides AI messaging for personalized outreach. Clay is SOC 2 Type II compliant and offers GDPR and CCPA support. Users can explore resources like Clay University, GTM blog, integrations, templates, and Chrome Extension.
Source:https://www.clay.com/
Numerous.ai
Numerous.ai offers a user-friendly and cost-effective solution for integrating ChatGPT into Google Sheets and Excel, eliminating the need for API keys. The tool allows for easy implementation with a simple =AI function, provides team collaboration features, and offers US-based support. It can assist in generating digital marketing content, summarizing and categorizing text, testing AI prototypes, and creating various types of content. Users such as content creators, digital marketers, students, researchers, and product engineers can benefit from Numerous.ai's capabilities.
Source:https://numerous.ai/
Julius AI | Your AI Data Analyst
Julius AI offers AI-powered data analysis services to over 1,000,000 users worldwide. Users can chat with their files to quickly gain expert-level insights, generate visualizations, automate data cleaning, and more. The platform supports various data file formats, including spreadsheets, Google Sheets, and Postgres databases. Students and educators can enjoy a 50% discount on subscriptions. Julius AI ensures strict data privacy policies and offers a free plan with limitations on monthly AI model usage.
Source:https://julius.ai/
GPT-4o System Card
OpenAI has released GPT-4o, an autoregressive omni model capable of processing any combination of text, audio, image, and video inputs and generating text, audio, and image outputs. The model was trained using data up to October 2023, sourced from a variety of materials including web data, code and math, and multimodal data. Prior to deployment, OpenAI carried out extensive safety evaluations and mitigations, including external red teaming and frontier risk evaluations. The model was assessed for potential risks such as speaker identification, unauthorized voice generation, and the generation of copyrighted content. Safeguards have been implemented at both the model and system levels to mitigate these risks.
NFL to roll out facial authentication software league-wide
The National Football League (NFL) is set to implement facial authentication technology across all 32 of its stadiums. The move comes after the NFL signed a contract with Wicket, a company that uses facial scans to verify identities at event venues. The technology will be used to secure and streamline entry for media, officials, staff, and guests into restricted areas. Wicket's technology was tested in a few NFL stadiums last year and will be rolled out league-wide starting August 8. Despite concerns from privacy advocates about the use of facial recognition technology, Wicket claims its technology has 99.7% accuracy and enhances security by preventing the use of fraudulent credentials.
OpenAI reveals prototype of AI powered search engine, SearchGPT
OpenAI has unveiled a prototype of an AI-powered search engine, SearchGPT, aimed at providing users with quick, accurate answers sourced from the web. The tool is designed to counter the issue of AI-generated noise online, which has reduced the effectiveness of SEO-driven search engines like Google. SearchGPT combines OpenAI's AI models with real-time web data and includes features that attribute web-based sources prominently. The company plans to integrate the most useful aspects of SearchGPT into ChatGPT in the future. OpenAI is currently testing SearchGPT with a small group of users and publishers.
At the Olympics, AI is watching you
In preparation for the Paris 2024 Olympic Games, the city has implemented a new type of AI to enhance security measures. The AI system, first tested in March at two Depeche Mode concerts, scans CCTV footage from transport stations to identify potential threats. The technology is designed to handle the vast amount of surveillance footage that is too extensive for human operators to monitor. The AI is programmed to detect eight specific "events," such as crowd surges, large groups, abandoned objects, weapons, or a person falling. Despite the increased security, the system has sparked controversy, with critics arguing it poses risks to fundamental rights.
Union game performers strike over AI voice and motion-capture training
SAG-AFTRA, the union representing game performers, has called for a strike over concerns about the use of their performances for AI training. The strike, which affects over 160,000 members, will impact games produced by companies such as Disney, Electronic Arts, Blizzard Activision, Take-Two, and WB Games. The union's chief negotiator, Duncan Crabtree-Ireland, has expressed disappointment at the studios' lack of progress in addressing these concerns. The main issue revolves around on-camera performers, including motion capture performers, whose work is being used without their informed consent and fair compensation. This strike follows a similar one in 2016, which lasted 340 days and focused on long-term royalties.
YouTube creators surprised to find Apple and others trained AI on their videos
Major tech companies including Apple, Salesforce, and Anthropic have reportedly trained their AI models on tens of thousands of YouTube videos without the creators' consent, potentially violating YouTube's terms. The training data was sourced from "the Pile," a collection by nonprofit EleutherAI, which includes YouTube captions from over 173,536 videos across more than 48,000 channels. Notably, Apple clarified that its model OpenELM, known to have used the Pile, was created for research purposes and does not power consumer-facing AI features on Apple devices. The use of such data has been a subject of multiple lawsuits, with defendants arguing that this kind of scraping is fair use.
X is training Grok AI on your data—here’s how to stop it
Elon Musk's social media platform, X, is training its AI chatbot, Grok, on user data, which is an opt-out feature rather than opt-in. Users discovered this when they noticed the opt-out checkbox in X's privacy settings, causing some to express their dissatisfaction. X's privacy policy has permitted this data usage since September 2023. Users can disable this feature on the mobile web, and X plans to introduce this option to its apps soon. Similar data usage practices are common among other companies, such as Meta.
US, UK ink AI pact modeled on intel sharing agreements
US and UK have signed a groundbreaking AI agreement to collaborate on assessing and regulating risks from emerging AI models. The pact involves sharing technical knowledge and talent to address potential threats from new technologies. The agreement enables the UK's AI Safety Institute and its US counterpart to exchange expertise and evaluate private AI models. The partnership is modeled after the intelligence-sharing agreement between the UK's GCHQ and the US NSA. The UK aims to lead in AI safety without immediate broad regulation, contrasting with other regions like the EU and the US. The UK's AISI, chaired by Ian Hogarth, is testing AI models for risks with the support of tech companies like Google DeepMind and Meta. The collaboration between the US and UK is seen as crucial in understanding and addressing the risks posed by AI.
Trickle rescues your screenshot chaos with AI.
Trickle AI introduces AI solution to organize screenshot chaos. Apple's Q3 FY23 Income Statement reveals key financial figures: Total Revenue of $81.8B, Gross Profit of $36.4B with a 45% margin, Operating Profit of $23.0B, and Net Profit of $19.9B. Major revenue sources include iPhone, Services, and MacBook/other products. Expenses consist of Cost of revenue and Operating expenses. The period covered is Q3 FY23, ending June 2023. Source: appeconomyinsights.com⎮Trickle AI.
Source:https://www.trickle.so/
DatologyAI is building tech to automatically curate AI training datasets | TechCrunch
DatologyAI, founded by Ari Morcos, is developing technology to automate the curation of AI training datasets, aiming to streamline the data preparation process for AI model training. The platform can identify crucial data, augment datasets, and optimize batch processing during training. Despite the potential of automated curation, some experts remain skeptical due to past issues with biased or toxic datasets. DatologyAI has received investments from prominent figures in the tech and AI industry, indicating confidence in the company's approach. The startup plans to expand its team and offerings in the near future.
This startup is using AI to discover new materials | TechCrunch
Orbital Materials, a startup founded by former DeepMind researcher Jonathan Godwin, is utilizing AI to discover new physical materials through their GenAI technology. Their AI-powered platform, including the proprietary model Linus, aims to revolutionize the materials science industry by accelerating the process of material discovery and development. With recent funding of $16 million in a Series A round, Orbital Materials plans to expand its data science and wet lab teams to further advance their innovative approach to materials research.
Abridge Emerges as a Healthcare AI Leader, Raising $150M
Abridge, a leader in generative AI for clinical documentation, secured a $150 million Series C investment led by Lightspeed Venture Partners. The funding will support their mission to improve healthcare by reducing the burden of documentation for clinicians. Abridge also announced a partnership with Yale New Haven Health System to provide thousands of clinicians with access to their AI platform. The company's innovative technology, including automatic speech recognition and evidence extraction, is setting new standards in healthcare AI. With this investment, Abridge plans to further advance their research and development efforts to enhance patient outcomes.
Microsoft's deal with Mistral AI faces EU scrutiny
Microsoft's investment in Mistral AI is facing scrutiny in the EU, with lawmakers calling for an antitrust investigation due to concerns of tech giant's growing power. This comes after Microsoft's previous partnership with OpenAI was already under review. Mistral's lobbying efforts for AI Act exemptions have raised questions about its motives. The deal has sparked criticism in Brussels, with some MEPs questioning the concentration of power and money. The French government and companies involved have not commented on the matter.
Tyler Perry Puts $800M Studio Expansion on Hold After Seeing OpenAI’s Sora: “Jobs Are Going to Be Lost”
Tyler Perry has put his $800 million studio expansion plans on hold after witnessing the capabilities of OpenAI's Sora, a text-to-video model. Perry expressed concerns about the impact of AI on jobs in the entertainment industry, emphasizing the need for regulations to protect workers. He highlighted the potential of AI to revolutionize production processes but also acknowledged the threat it poses to traditional job roles. Perry mentioned using AI in two upcoming films to streamline production. He called for a collective industry effort to address the challenges posed by rapidly advancing AI technology.
Apple to Wind Down Electric Car Effort After Decadelong Odyssey
Apple is discontinuing its decade-long project to build an electric car, surprising nearly 2,000 employees working on the initiative. The decision was internally disclosed by Chief Operating Officer Jeff Williams and Kevin Lynch, a vice president overseeing the effort.