You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
Will Nvidia and TSM use chiplet technology to make customized AI/ML chips rather than rely on a one size fits all AI/ML chips. Is TSM the leader in chiplet technology? Could this lower the cost of applying AI/ML in different areas? Will there be partial specialization applied to keep the costs down? Thanks