Dropbox Introduced Dash & $50M AI Initiative
Plus: Build powerful apps with AI Workbook. Microsoft proposes a smaller, faster coding LLM.
Hello Engineering Leaders and AI Enthusiasts!
Welcome to the 47th edition of The AI Edge newsletter. This edition brings you
“Dropbox Introduced Dash & $50M AI Initiative”.
And a big thanks to all our incredible readers! 😊
In today’s edition:
💥 Dropbox Introduced Dash & $50M AI Initiative
💡 AI Workbook helps you build powerful apps
📈 Microsoft Research proposes a smaller, faster coding LLM
📚 Knowledge Nugget: How RLHF actually works by
Let’s go!
Dropbox Introduced Dash & $50M AI Initiative
Dropbox has launched two new AI-powered tools, Dropbox Dash and Dropbox AI, to enhance productivity and improve knowledge work. These tools leverage recent advancements in AI and ML to provide personalized assistance, answer questions, offer insights on content, and facilitate efficient content search within the Dropbox platform.
Meet Dropbox Dash: The AI-powered universal search tool for work
A quick look at Dropbox AI
Why does this matter?
The AI initiatives by Dropbox can have significant implications for knowledge work and productivity. It has the potential to automate repetitive tasks, provide personalized assistance, and offer valuable insights, thereby saving time and enabling users to focus on more high-value work. This development reflects Dropbox's belief in the transformative potential of AI.
AI Workbook helps you build powerful apps
AI Workbook is a notebook interface for using LLMs, images, and audio models all in one place! That will help you to build powerful applications and workflows in a single environment.
What you can do with AI Workbook:
Combine text, image, and audio models in one place.
Customize AI models using your data.
Share your work & collaborate.
Why does this matter?
This innovative platform can potentially reshape how AI applications are built, customized, and shared. One of the key reasons why AI Workbook matters is its ability to combine multiple AI modalities seamlessly. With this, developers can create more comprehensive and sophisticated AI applications, opening up new possibilities for automation, analysis, and decision-making.
Microsoft Research proposes a smaller, faster coding LLM
Microsoft Research has proposed a new LLM for code in its paper Textbooks Are All You Need. Significantly smaller in size than competing models, phi-1 is a Transformer-based model with 1.3B parameters, trained for 4 days on 8 A100s, using a selection of “textbook quality" data both synthetically generated (with GPT-3.5) and filtered from web sources, and finetuned on “textbook-exercise-like” data.
The model surpasses almost all open-source models on coding benchmarks, such as HumanEval and MBPP (Mostly Basic Python Programs), despite being 10x smaller in model size and 100x smaller in dataset size.
Why does this matter?
This work shows how high-quality data can improve the learning efficiency of LLMs’ and their proficiency in code-generation tasks while dramatically reducing the dataset size and training compute. It also jumps on the emerging trend of using existing LLMs to synthesize data for training new generations of LLMs.
Knowledge Nugget: How RLHF actually works
How does reinforcement learning from human feedback (RLHF) work– this has been one of the most asked questions about AI. In this insightful article,
discusses the working principles of RLHF and highlights two conditions for RLHF to be effective.He starts with the starting point for RLHF and goes on to explain the process of collecting preference data and guiding the model through RL optimization. The article also mentions ongoing research, challenges, and future directions in RLHF.
Why does this matter?
The understanding of workings of RLHF can inform researchers and developers in designing and implementing RLHF algorithms more effectively. Moreover, it can help them address RLHF’s limitations and devise strategies to overcome them, enabling building robust AI systems.
What Else Is Happening ❗
🎥 Marvel uses AI to make the opening credits of Secret Invasion (Link)
🚀 Google Cloud launches AI-powered anti-money laundering tool for banks (Link)
🤳 Meta expands reels ads to Instagram, tests AI features (Link)
🔊 Wimbledon to use AI commentary at 2023 Championships (Link)
💡 Microsoft introduces Bing AI in Skype (Link)
🧠 Harness releases generative AI assistant to help increase developer efficiency (Link)
🛠️ Trending Tools
Upword AI: Research tool for knowledge workers. Version 2.0: process docs, AI research copilot, organize content, listen & share.
Otter AI: Collaborative AI for meetings. Answers questions, collaborates, generates content based on meeting data.
Caden AI: Streamline customer service! Let Caden read messages, generate responses, or craft professional replies.
Rebecc AI: Evaluates ideas on a 5-star scale, refines concepts, and revolutionizes ideation. Quick insights for shaping ideas.
Monitup: Boost productivity with AI and daily activity log software.
Webbrain AI: Innovating website creation with powerful AI. Craft stunning sites, explore intuitive tools, seamless development.
Chapters AI: Chat client for GPT that branches paragraphs. Ideal for building projects and exploring component parts.
SalesGPT: Powerful AI trained on Bravado's sales database. Guides sales professionals through revenue challenges.
That's all for now!
Subscribe to The AI Edge and join the impressive list of readers that includes professionals from Moody’s, Vonage, Voya, WEHI, Cox, INSEAD, and other reputable organizations.
Thanks for reading, and see you tomorrow.