First AI Chip That Learns and Infers
The first AI chip that learns and infers has been developed, combining FeCAPs and memristors for efficient, adaptive edge computing.
According to Robin Mitchell, one of the main obstacles to progress in artificial intelligence is physical, not algorithmic.
The current approach to AI systems relies on large amounts of specialized silicon, high energy consumption, and complex cooling systems, which is unsustainable.
- Data centers strain power grids
- Industrial-scale cooling is required, often using water-intensive processes
Training AI processes is the most energy-intensive part, and if AI continues to scale at its current pace, it could lead to serious resource shortages.
could see serious resource shortages
Author summary: AI chip learns and infers efficiently.
More News
- Benz Hui Dies At 76 from Cancer-Related Organ Failure
- Accessible, low-cost biomedical innovations at Tufts’ research lab - The Tufts Daily
- Park Washroom Enhancement Program
- Mark Rutte’s Dutch liberals were dominant for years. Now the party is bleeding support.
- Australian government bans Dutch volleyball rapist
- Antifa Terrorism Charges: A Victory for Law Enforcement
- Closed-Loop Insulin System Improves Glucose Control in Pregnant Women with Type 1 Diabetes: JAMA
- Utica outlasts Warren Cousino; Lakeview topples rival Lake Shore; Chippewa Valley, Warren Mott also record lopsided wins
- Landmark New Report Shows Canadian Lobbyists Represent Major Polluters Alongside Environmental Groups