OpenAI's AI just turbocharged AMD's chip designs in record time
Original headline: OpenAI says its models delivered a year's worth of chip optimization on top of AMD's own engineering — the-decoder.com
More in AI and machine learning news
OpenAI, in collaboration with AMD, has deployed its AI models to perform chip optimization at an unprecedented pace. The company claims its GPT-6.2 Astra model achieved up to a 14% improvement in power, performance, and chip area (PPA) on AMD's flagship GPU intellectual property in under five months. This leap forward represents roughly a full year's worth of progress on top of the deep optimizations AMD's own engineering teams had already made over years, with AMD already incorporating some of these AI-generated improvements into a design now being manufactured.
The significance is a fundamental shift in semiconductor design velocity. OpenAI's method also accelerated the design of its own custom inference chip, 'Jalapeño,' and delivered top results on open RISC-V cores. By partnering with Synopsys on 'GPT-Synopsys,' OpenAI aims to democratize AI-assisted chip design. Meanwhile, competitor Anthropic reports its 'Mythos' model completed a complex GPU prompt-processing task in three days—a task that previously took six to twelve months. The takeaway is clear: generative AI is no longer just generating text and images; it's now generating better, faster, and more efficient hardware, permanently altering the cadence of innovation in the semiconductor industry.
Read the original at the-decoder.com
Satirical commentary on real tech news, summarised with AI. Read the original source for the facts. How ZeroDaily works
