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 biggest challenges in artificial intelligence is not algorithmic, but physical. Current AI systems require large amounts of specialized silicon, consume a lot of energy, and need complicated cooling systems.
This approach is unsustainable, and if AI continues to grow at its current rate, it could lead to serious resource shortages. Data centers already put a strain on power grids, and the heat generated requires industrial-scale cooling, which often uses water-intensive processes that have their own environmental and economic issues.
Training AI models is the most energy-intensive process, highlighting the need for more efficient solutions.
AI continues to scale at its current pace, and could see serious resource shortages.
By Robin Mitchell.
Author's summary: New AI chip merges FeCAPs and memristors.
More News
- Telcos and ISPs: The time to prioritise cyber security is now
- Former Rutgers Rowers Reunite to Help Families of Fallen First Responders
- 'Bargain' Liverpool move for Bundesliga livewire mooted and Ekitike is the secret weapon - report
- Is your organization ready for the future of manufacturing? Gartner reveals statistics
- Pre-school praised for 'working hard to improve' with exemplary Ofsted report
- Code stroke of large vessel occlusion at triage improves candidate selection and workflow efficiency of endovascular thrombectomy | 2 Minute Medicine
- Policy Brief: Internet Shutdowns - Internet Society