Sustainable by design: Innovating for energy efficiency in AI, part 2 | The Microsoft Cloud Blog
Explore examples of how Microsoft is bringing promising efficiency research out of the lab and into commercial operations.| The Microsoft Cloud Blog
Explore examples of how Microsoft is bringing promising efficiency research out of the lab and into commercial operations.| The Microsoft Cloud Blog
The future of generative AI could rely on smaller language models for every application an enterprise uses, models that would be both more nimble and customizable — and more secure.| Computerworld
By default powerful ML systems will have dangerous capabilities (such as hacking) and may not do what their operators want. Frontier AI labs should design and modify their systems to be less dangerous and more controllable. In particular, labs should:| ailabwatch.org
When a dangerous model is deployed, it will pose misalignment and misuse risks. Even before dangerous models exist, deploying models on dangerous paths can accelerate and diffuse progress toward dangerous models.| ailabwatch.org
Learn how Alex bought a set of chisels by monitoring the price using HTML scraping and a cron job with OpenFaaS. You can apply the same technique for things you’d like to buy or monitor.| OpenFaaS - Serverless Functions Made Simple
I explore the competition between two data powerhouses and what it says about where enterprise AI is headed.| Foundation Capital
As part of the Phi-3 launch Microsoft has released optimized ONNX models as detailed in ONNX Runtime supports Phi-3 mini models across platforms and devices and published the models on HuggingFace 🤗 at Phi-3 Mini-4K-Instruct ONNX models for consumption in for example the ONNX Runtime GenAI. This makes it very easy to run this model locally in just a few lines of C# as I’ll show in this blog post.| nietras.com
