←Search results

Search “OpenAI”

By relevance · Total 3708 results Clear search

Friday

Thursday

  5. Cailian Press

    Six tech giants' executives gather at White House lunch: Where does the AI safety debate go from here?

    Six tech executives—Meta's Zuckerberg, Anthropic's Amodei, OpenAI's Brockman, Google's Pichai, Palantir's Karp, and Nvidia's Jensen Huang—are expected to gather at the White House on September 29 local time for a lunch also attended by Trump and House Speaker Johnson. The debate in the U.S. tech world over AI risks has split into two camps; Amodei previously wrote an article calling for "putting the brakes on the pace" of frontier AI development. In the short term, it is almost impossible for the U.S. Congress to pass any AI-related bill.

    Six tech executives—Meta's Zuckerberg, Anthropic's Amodei, OpenAI's Brockman, Google's Pichai, Palantir's Karp, and Nvidia's Jensen Huang—are expected to gather at the White House on September 29 local time for a lunch also attended by Trump and House Speaker Johnson. The debate in the U.S. tech world over AI risks has split into two camps; Amodei previously wrote an article calling for "putting the brakes on the pace" of frontier AI development. In the short term, it is almost impossible for th

    Matched fieldsSummary、Body

    #chinese#Industry news#OpenAI#Anthropic
  7. A6 a6-051 · Cerebras：Blog

    Cerebras: Latency Debt Is Dragging Down AI Experience, GPUs Have Hit a Bottleneck

    Cerebras proposes the concept of "latency debt": model parameters have grown from 340M in BERT-Large to over 1T in Kimi-k2, the share of inference tokens has risen from under 10% to over 50%, context has expanded from about 1,000 tokens in GPT-2 to 10 million tokens, input length has doubled every year and output has tripled, causing latency to accumulate continuously. Cerebras says its Wafer-Scale Engine (WSE) keeps data on-chip, eliminating the memory bandwidth bottleneck of GPU inference, and OpenAI has already purchased 750 MW of compute from it.

    arameters have grown from 340M in BERT-Large to over 1T in Kimi-k2, the share of inference tokens has risen from under 10% to over 50%, context has expanded from about 1,000 tokens in GPT-2 to 10 million tokens, input length has doubled every year and output has tripled, causing latency to accumulate continuously. Cerebras says its Wafer-Scale Engine (WSE) keeps data on-chip, eliminating the memory bandwidth bottleneck of GPU inference, and OpenAI has already purchased 750 MW of compute from it.

    Matched fieldsSummary、Body

    #a6-editorial#Industry news#English text is being prepared.#Reasoning

Wednesday