Startup EnCharge AI announced on Feb 13 that it had secured over $100 million in a Series B funding round led by Tiger Global. The funding aims to introduce more efficient and cost-effective AI chips to the market. The company specializes in developing analog chips integrated into semiconductors utilized in storage, specifically focusing on in-memory chips tailored for inference, where AI models are applied rather than trained.
Contrary to the common practice of housing AI inference chips in large server clusters within data centers, EnCharge AI's chips are tailored for edge computing, serving user-facing devices like laptops. By incorporating analog processing into memory chips, their accelerators can execute AI tasks with up to 20 times less energy consumption than some leading AI chips.
Naveen Verma, CEO of EnCharge AI, highlighted the importance of efficient chips for battery-powered devices like laptops and smartphones. He emphasized the advantages of analog chips within memory semiconductors in addressing cost, sustainability, privacy, and security concerns.
Groq and Cerebras, companies developing specialized chips for AI inference, were mentioned alongside Tiger Global as investors. Additionally, the funding round attracted investments from Samsung Electronics' VC arm and HH-CTBC, a partnership between Taiwan's Foxconn and CTBC Venture Capital.