- Source : https://www.deeplearning.ai/the-batch/issue-284/ - Auteur: [[https://www.facebook.com/DeepLearningAIHQ/]] - Notes liées : *** # Résumé The U.S. is expanding export restrictions on AI technology, creating a tiered system that limits access for countries like China and Russia. These new rules aim to keep advanced AI development concentrated among close allies while encouraging the growth of clean energy-powered AI infrastructure. As AI technology evolves, the demand for skilled AI Product Managers will increase, requiring a unique set of technical and management skills. # Highlights - DeepSeek Ups the Open Weights Ante - **Mixture of experts (MoE) basics:** The MoE architecture uses different subsets of its parameters to process different inputs. Each MoE layer contains a group of neural networks, or experts, preceded by a gating module that learns to choose which one(s) to use based on the input. In this way, different experts learn to specialize in different types of examples. Because not all parameters are used to produce any given output, the network uses less energy and runs faster than models of similar size that use all parameters to process every input. - **How it works:** DeepSeek-V3 is a mixture-of-experts (MoE) transformer that comprises 671 billion parameters, of which 37 billion are active at any moment. The team trained the model in 2.79 million GPU hours — less than 1/10 the [time required to train Llama 3.1 405B](https://info.deeplearning.ai/e3t/Ctc/LX+113/cJhC404/VVDS5m4T6qf8W427Ds551gWLTW46ZLVZ5qSGfcN4xH9dY3qgyTW7Y8-PT6lZ3l8W4HZ3tr729ClLW935Ks498jK8WW6Cpm2f4bq26RW7GtCyN6lqPkGN97q92ns8MKGW1LtnMh7kvBhTW2D4x7F6gY8DsN6LxsSrg1PbCN3Vdp0CV4HQhW3Bv-4h3KqfwhN5jphRqJCyy9VlDbmb2pgxH6W48_SQd7SKKDfN9jX4rxVQmrzW80Y9Fs4XR_g4W3-9n_K7wKGV9V4nGPl5kd8XPW7NMjsB1vws0-W5RNSs02nwN1sW7swJ636y0Mc4W4QzSrm5LsTBqW4h-fN73yNG_KW1x6_PX21pqwkW8hgxgZ19z7w-N6xNv5qKTwqmW6x8nZX3R7r3Lf3Bwftg04?ref=dl-staging-website.ghost.io), which DeepSeek-V3 substantially outperforms — at an extraordinarily low cost of $5.6 million. - **Why it matters:** Open models continue to challenge closed models, giving developers high-quality options that they can modify and deploy at will. But the larger story is DeepSeek-V3’s shockingly low training cost. The team doesn’t explain precisely how the model achieves outstanding performance with such a low processing budget. - U.S. Moves to Expand AI Export Restrictions - **What’s new:** The Biden administration, which will transition to leadership under incoming President Trump next week, issued new [rules](https://info.deeplearning.ai/e3t/Ctc/LX+113/cJhC404/VVDS5m4T6qf8W427Ds551gWLTW46ZLVZ5qSGfcN4xH9cs5nR32W50kH_H6lZ3nPW2CxHnd3xC8R7W85GfFd9lRg2xW3xN8RX6vFrPgW1BL9Yq4HSPD-W6g6Gzl8FbzqNW74l9Gf5NB3nQV6-7pr1fmPs4W3zwdhK72N2M4W6szqF43zDkwHW6J8hQ98fZMNDVNcPyW6pVlWBW3mbmBj522-sqW4HKfGy6p22gFMkyg6wjt61NVk4Qpx5sLswvW19Ctfp8Db2ZHW6JNqRv1tS_WYW7zCJrM6kKNnXW58sM9-6sxTsHW8q5l2x9d7y6NW4X5m9r662TkzW6MM9T865-LZmW99xxmD7pYxFyMRwmVCl6XHqW5yC_Zg1XVnPwW77MvZk3skDS6W85PVrT6KnrBDW21MNJ612gFqnVcSMM22H2Zq7W3dNJnC5wBg0WW7FMZ5q1nDKXNW54V-Rc1t_HXXf3xP8VK04?ref=dl-staging-website.ghost.io) that restrict exports of AI chips and models to most countries beyond a select group of close allies. The rules, which are not yet final, would create a three-tier system that limits exports to a number of close allies and blocks access entirely to China, Iran, North Korea, Russia, and others. They also would introduce the U.S.’ first-ever restrictions on exporting closed weights for large AI models. - **How it works:** The restrictions were announced shortly after a [leak](https://info.deeplearning.ai/e3t/Ctc/LX+113/cJhC404/VVDS5m4T6qf8W427Ds551gWLTW46ZLVZ5qSGfcN4xH9dY3qgyTW7Y8-PT6lZ3kYW1Spmy-6WjWPCW2TG1R24KbTXhW47ph9X97mbWhN4WV55g6Pt2BW2LVxj05zHBhsV65ddR7tqCS2W2Fd3T5759xDHW8RVQRb1YjS3_W52q69j5f21c3W52X3Hq4s5CxBW19qTJx4YGH5sVMZsbK2NstcnW17jS-T5v7qWlW95T7JZ6FTn_KW5tr1xG1QfWCLN2xPKKj63jqyW13T4s16rC8gSW7BRBYb5Cqs8dW2wp8X47LWxVyW6Q0dw08pxNt7N886Jt3Qfhr-W5Rhp9F1GDY0LN8Y_tT9GgC5xW5vbk_T2-ZprKN5WyKzKWcRP4W3Q1RX_30k8LVf2k72Nv04?ref=dl-staging-website.ghost.io) reached the press. A public comment period of 120 days will enable the incoming U.S. Presidential administration to consider input from the business and diplomatic communities and modify the rules before they take effect. The rules are scheduled to take effect in one year. - **Why it matters:** Protecting the United States’ advantages in high tech has been a rising priority for the White House over the past decade. The earlier export restrictions forced many Chinese AI developers to rely on less-powerful hardware. The new limits are likely to have a far broader impact. They could force developers in the Tier 2 and Tier 3 countries to build less resource-intensive models and lead them to collaborate more closely with each other, reducing the value of U.S.-made technology worldwide. They could hurt U.S. chip vendors, which have [warned](https://info.deeplearning.ai/e3t/Ctc/LX+113/cJhC404/VVDS5m4T6qf8W427Ds551gWLTW46ZLVZ5qSGfcN4xH9dF3qgyTW7lCdLW6lZ3pjW2Gc-GB8D0XgnW3kQvhz8Bf__6N6y-6FNZRy73W6yRFn81Wg2wLN6M_4p0qfWsTW7jtm1g20C34kW7WXWTk1qksxgW453s0x8BSv8bW7RcFGM7LZjZsN3zQ78lZ5lrgW3D-GB28KjqJTVQGNNx6bxgZbW5TBk3S6RWNy6W5Q9F2J6zp5RNW3Rgvwt4_yMgcW1MTTkF9g4Xs3W1cJmj-7Rmz-qW7CDvdF48lMLrW6ZC9LM7zh59yV1_dRp5-4QJjN2wNFn-VKLpwW2BC6hd5FHhz0W7BKpdm2nGHmTW6yX-rf66GhpLf23wdZq04?ref=dl-staging-website.ghost.io) that the rules could weaken U.S. competitiveness in the global economy. They could also force companies that are building huge data centers to process AI calculations to [reconsider](https://info.deeplearning.ai/e3t/Ctc/LX+113/cJhC404/VVDS5m4T6qf8W427Ds551gWLTW46ZLVZ5qSGfcN4xH9fd3qgyTW8wLKSR6lZ3kYV5Vvd03zNf7fVYZ1R42xZL4qW1M-2z08M_YHgN1F1dyFZw3MJW6XrfsY927FlGW8g1y7N4WVWbSW1qcfZ23TVykZW5dkm478TGnqsW6mpylD7yNFJqW12g6mn78jHX8W2t9GXZ4xKCRzW7yp45P1fQMSkW8S3LR63zd92BW7mf2gr489MCZW1f2bfV4CG_xnW7zZZMt5Q-Qt1VF0F8_5bNJK_W8TXP0z57pPCmW2ZYMgB7lSXQZW94jpCb1D-Gh3W2kqS247W9xmQN3ZHxm86hp4jW56bVqf81kZTwW84cwfc6FD8N1W5gf_c52M2815W1XLlbQ5kMCf6W6Lc3Z48RbLHdN6YXCPwr2-C_f9bD4Lz04?ref=dl-staging-website.ghost.io) their plans. - AI Supercomputer on Your Desk - **What’s new:** [Project Digits](https://info.deeplearning.ai/e3t/Ctc/LX+113/cJhC404/VVDS5m4T6qf8W427Ds551gWLTW46ZLVZ5qSGfcN4xH9dF3qgyTW7lCdLW6lZ3ngW6M9rtw1KpkCLW35nzTZ6kxhJZN92X3KDhcBCrW87x7xs3cC5lnW5h9Fdy5mfNL_W295Nnm8bb016W2WP-TP5KBrZvW20GzsJ5yR4szN447cpl-b_55N7LkW7QqnQnzW5yTTHt7zX7SPW6xqPW14K0t2TW8BJxrf8RNcS4W29XMn24zCDXbW43q1tK1gctl7W5SmzpW8Wh3zGW7GR2Jt46vnMVW7k8hXg87m1cnN2Jn2kfKFYV2W4gbK1d4ljfxzW8qlSBj6rk-JYW3-V1BC3SZd9WW3rXrPb19VkLfW1WZBB26_g00Ff5q8rVv04?ref=dl-staging-website.ghost.io) is a personal supercomputer intended to help developers fine-tune and run large models locally. Project Digits, which is small enough to hold in one hand, will be available in May, starting at $3000. - **How it works:** Project Digits is designed to run models of up to 200 billion parameters — roughly five times the size that fits comfortably on typical consumer hardware — provided they’re quantized to 4 bits of precision. Two units can be connected to run models such as Meta’s Llama 3.1 405B. Complete specifications are not yet available. - **Why it matters:** It’s common to train models on Nvidia A100 or H100 GPUs, which come with a price tag of at least $8,000 or $20,000 respectively, along with 40 gigabytes to 80 gigabytes of memory. These hefty requirements push many developers to buy access to computing infrastructure from a cloud provider. Coming in at $3,000 with 128 gigabytes of memory, Project Digits is designed to empower machine learning engineers to train and run larger models on their own machines.