Daily AI Newsletter: Meta’s battle with ChatGPT begins now

In today’s email:

🔥 Meta Is Already Training a More Powerful Successor to Llama 3
👨🏽‍⚕️ A benchmark for testing generative AI on health tasks
👀 How cheap, outsourced labor in Africa is shaping AI English
🧰 9 new AI-powered tools and resources. Make sure to check the online version for the full list of tools.

Meta’s battle with ChatGPT begins now

Meta is intensifying its efforts in the AI space with the integration of its AI assistant across its family of apps including Instagram, WhatsApp, and Facebook. Mark Zuckerberg recently announced the release of Llama 3, Meta's latest AI model, which claims to outperform existing models on various benchmarks, especially in coding tasks. This new model is now available in two versions, with a larger, multimodal version expected soon. Zuckerberg's vision is for Meta AI to become the most utilized and intelligent AI assistant globally.

The Meta AI assistant now features enhanced capabilities such as real-time image generation and the integration of real-time search results from Google and Bing within its interface. Initially available only in the US, the assistant is now expanding to other English-speaking countries. The assistant aims to be a comprehensive tool by also generating suggestions to help users understand the capabilities of a general-purpose chatbot.

Moreover, the development of Llama 3 reflects a significant scale-up, with the model now trained on over 15 trillion tokens, a significant increase from its predecessor, Llama 2. Zuckerberg is considering making the upcoming 400-billion-parameter version open-source, emphasizing safety and transparency in AI development. Meanwhile, Meta is preparing for future iterations, with Llama 4 and 5 already in conceptual stages, highlighting the company's commitment to maintaining a lead in the AI technology race.

AI-Powered Productivity: Transform Your Day

Revolutionize your workday with the power of ChatGPT! Dive into HubSpot’s guide to discover how AI can elevate your productivity and creativity. Learn to automate tasks, enhance decision-making, and foster innovation, all through the capabilities of ChatGPT. Download the offer for actionable insights and tips on harnessing AI for a more efficient, impactful professional life.

Efficiency Unlocked: Enhance productivity with daily ChatGPT integrations
Everyday Solutions: ChatGPT for streamlined email and project management
Practical Insights: Guidelines for effective ChatGPT application in work
Ethical Guidelines: Best practices for responsible ChatGPT use
Future Ready: Skill enhancement and learning with AI tools

Download guide for Free

Hugging Face releases a benchmark for testing generative AI on health tasks

Hugging Face has introduced Open Medical-LLM, a new benchmark designed to evaluate generative AI models in healthcare tasks. This benchmark, developed in collaboration with Open Life Science AI and the University of Edinburgh, combines various test sets from the medical field, including MedQA and PubMedQA, to assess AI capabilities in medical reasoning and knowledge. Open Medical-LLM uses both multiple choice and open-ended questions drawn from medical licensing exams and biology tests, aiming to help identify the strengths and weaknesses of different AI approaches in healthcare.

Despite the potential benefits, there are concerns about the practicality of deploying these AI models in real-world medical settings. Critics argue that the controlled conditions of these tests do not accurately reflect the complexities of clinical practice. This sentiment is echoed by medical professionals and researchers who emphasize the need for extensive testing beyond initial benchmarks to ensure AI tools' effectiveness and safety in actual healthcare environments.

The skepticism around generative AI in healthcare is supported by past experiences, such as Google's attempt to implement an AI system for detecting diabetic retinopathy in Thailand, which faced challenges in real-world application despite high accuracy in test environments. The FDA has yet to approve any AI medical devices that rely solely on generative models, highlighting the difficulty in predicting how these tools perform in hospitals and their impact over time. This underscores the importance of thorough and careful testing beyond initial AI benchmarks like Open Medical-LLM.

Nothing’s ear (a): First ChatGPT Integrated Earbud

Nothing's latest product release includes the Nothing Ear and Nothing Ear (a), priced at $149 and $99 respectively. The design remains largely unchanged from previous models, featuring the distinctive transparent elements first introduced in 2021. Both models have improved sound profiles and extended battery life, with the Nothing Ear boasting advanced driver systems and customizable sound settings through a companion app. The Ear (a) offers a new splash of color and is slightly less expensive while maintaining many of the same features as its counterpart.

Both earbud models come equipped with effective active noise cancellation, are capable of muffling up to 45dB of noise, and are dust and water-resistant to varying degrees. The Nothing Ear includes perks like wireless charging and a sturdier case, while the Ear (a) features a more compact case design inspired by every day pill packets. Both sets have improved voice call capabilities and AI voice isolation, enhancing the overall user experience.

The Nothing Ears competes well in audio quality against more established brands, offering a variety of connectivity options and a low-latency gaming mode at a competitive price. While they lack certain features like spatial audio found in products from Apple or Samsung, the sound quality and customization options are comparable. Additionally, Nothing plans to integrate ChatGPT into these devices, promising an enhanced interactive experience with AI in the future.