If you’re a gamer, and even if you’re not, you may have heard of Jensen Huang. He’s the CEO of Nvidia, the company behind some of the world’s most advanced chips and processors, originally intended for the gaming industry, but now powering the evolution of artificial intelligence. ChatGPT, ElevenLabs, OpenAI, Google, Microsoft and scores of other companies developing AI products are all powered by Nvidia servers and software.

In March, Huang gave an address to attendees at the GPU Technology Conference in Hong Kong, an annual gathering to discuss developments in the AI industry. During his speech, he demonstrated remarkable advances in computer graphics, drawing cheers and applause from the audience. But he also wheeled out Nvidia’s latest processor, a beast called “Blackwell,” powered by a superchip that can process up to 1.8 terabytes per second (that’s one trillion bytes, the equivalent of two hundred thousand copies of Webster’s Collegiate Dictionary).

What does it mean for AI to wield such computing power? Simply that generative AI, artificial intelligence that can learn from data it processes, will be able to digest massive amounts of information and learn at a rate unimaginable even a year ago. It is not unreasonable to expect a single Nvidia “GAI farm” of hundreds of Blackwell processors will be able to absorb and “understand” the entire history of Western civilization in the time it takes to drive to your local library and check out a copy of H.G. Wells’ The Outline of History.

If this impresses (and maybe scares) you, check out Huang’s presentation yourself, youtube.com/watch?v=Y2F8yisiS6E&t=258s. If nothing else, you’ll be amazed by the graphics!