We launched Telegram on August 14, 2013 – and immediately set to work on the next update. That's how it's been ever since. This page lists all major Telegram updates in chronological order.

Scientists have developed a new bio-synthetic coating for electronic components that could help merge AI with the human brain.

Live webcast of working @Neuralink device Friday 3pm Pacific https://t.co/PouLbrGzFU

TF-Coder is a program synthesis tool that helps you write TensorFlow code. Instead of coding a tricky tensor manipulation directly, you can just demonstrate it through an illustrative example, and TF-Coder provides the corresponding code automatically.

Tune in for the latest advancements in Elon Musk's Neuralink technology at 3pm PT/6pm ET.
00:00 Waiting for Broadcast to start
48:48 Neuralink Presentation
1:12:39 Q&A Session with Audience

A refined and improved digital notepad

Whenever I visit the Sahara I am struck by how sunny and hot it is and how clear the sky can be. Aside from a few oases there is little vegetation, and most of the world's largest desert is covered with rocks, sand and sand dunes.

Drew Houston, the founder of Dropbox said:

Researchers at Google have developed a new deep-learning model called BigBird that allows Transformer neural networks to process sequences up to 8x longer than previously possible. Networks based on this model achieved new state-of-the-art performance levels…

Drew Houston, the founder of Dropbox said:

Complete, end-to-end examples to learn how to use TensorFlow for ML beginners and experts. Try tutorials in Google Colab - no setup required.

AI can now code in any language without additional training.

Hivemind is a library to train large neural networks across the internet. Imagine training one huge transformer on thousands of computers from universities, companies, and volunteers.

We've used reinforcement learning from human feedback to train language models for summarization. The resulting models produce better summaries than 10x larger models trained only with supervised learning: https://t.co/Sk31d1CnTu