The World This Week in AI (29th July 2024)
OpenAI Launch GPT-4o Mini!
GPT-4o mini, a cost-efficient AI model aimed at making advanced AI capabilities more accessible. Priced at 15 cents per million input tokens and 60 cents per million output tokens, GPT-4o mini is over 60% cheaper than its predecessor, GPT-3.5 Turbo. Despite its affordability, GPT-4o mini excels in various tasks, including mathematical reasoning, coding, and multimodal reasoning, outperforming comparable models like Claude Haiku and Gemini Flash in benchmark tests. The model features an extended context window of 128K tokens and supports up to 16K output tokens per request, allowing it to generate detailed and contextually relevant responses.
One of the significant advantages of GPT-4o mini is its advanced safety measures. OpenAI has built in safety protocols from the start, utilizing techniques such as reinforcement learning with human feedback to ensure reliable and secure model responses. GPT-4o mini adheres strictly to guidelines, minimizing the risk of misuse. The model is available through various APIs, including the Assistants API, Chat Completions API, and Batch API, with fine-tuning options planned for the near future. It is accessible to Free, Plus, and Team users on ChatGPT, with availability for Enterprise users also on the horizon.
Microsoft and Lumen Partner to Advance AI and Digital Transformation?
Lumen Technologies and Microsoft have partnered to leverage the Microsoft Cloud for Lumen’s digital transformation.?
A Network for the AI Era
Microsoft is partnering with Lumen to enhance its network for next-gen AI applications like Azure OpenAI Service and Microsoft Copilot. Lumen’s Private Connectivity Fabric? will provide robust connectivity between Microsoft's data centers. Erin Chapple, Microsoft's VP, emphasized AI's transformative role and the need for reliable infrastructure. Lumen’s CEO, Kate Johnson, highlighted the importance of a powerful network for AI innovation, noting Lumen's unique routes and secure digital platform.
领英推荐
Accelerating Growth Through Innovation and Partnership
Lumen is transforming its operations using Microsoft’s cloud and AI technology to cut costs, eliminate legacy systems, and enhance services. Migrating to Azure and using Microsoft Entra for security, Lumen expects to improve cash flow by over $20 million in a year and boost customer experience. This builds on their long partnership, with Lumen leveraging Copilot for automation and enhancing Microsoft Teams.
Amazon's Prime Video Now Using AI to Recommend Movies and Shows
Amazon has started rolling out a new update to its streaming platform Prime Video, aimed to help users quickly find content they actually want to watch. The new features include an updated navigation bar and more personalized recommendations powered by generative AI.
The update includes a redesigned navigation bar at the top of the app, allowing users to switch between the home page, movies, TV shows, sports, and live TV sections. This bar also provides access to add-on subscriptions such as Paramount+ and Max. A new "Prime" tab has been introduced to display only content available with a Prime subscription, distinguishing it from rental or purchase-only content.
Additionally, the update introduces "Made for You" collections under the Movies and TV Shows sections, which are personalized recommendations powered by generative AI.?
Meta’s AI New Selfie Feature “Imagine Me”
Meta AI has just released the new Llama 3.1, in Llama 3.1 405B has a new generative AI model and it powers the selfie feature "Imagine Yourself" allows users to generate images based on a photo and a prompt, such as “Imagine me surfing” or “Imagine me on a beach vacation.” To use it, type “Imagine me” followed by your desired scenario, as long as it isn't NSFW.
Additionally, new editing tools in Meta AI let users modify images with prompts. Starting next month, an “Edit with AI” button will offer more fine-tuning options. Soon, users will also see shortcuts for sharing AI-generated images across Meta platforms.