index.html

<!DOCTYPE html>
<html lang="en">
  <head>
    <meta charset="UTF-8" />
    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
    <title>TypeSafe AI</title>
    <link rel="icon" type="image/svg+xml" href="./logo.svg" />
    <style>
      body {
        margin: 0;
        padding: 20px;
        max-width: 100%;
        box-sizing: border-box;
        font-family: Arial, sans-serif;
        line-height: 1.6;
      }
      .container {
        max-width: 600px;
        margin: 50px auto;
      }
      p {
        margin-bottom: 15px;
      }
      b {
        color: black;
      }
    </style>
  </head>
  <body>
    <div class="container">
      <p>
        The future of AI should be transformative—not incremental. Today,
        leading labs like OpenAI and Anthropic focus on LLMs optimized for human
        assistance. Yet, instead of unleashing a sci-fi vision of commoditized
        intelligence and economic revolution, we've settled for copilots and
        marginal productivity gains.
      </p>
      <p>
        Our team—led by Diogo Almeida (OpenAI, Google Brain, co-inventor of
        InstructGPT, RLHF, ChatGPT, and GPT-4), Erik Gafni (Ravel co-founder,
        multimodal AI), and Sasha Sheng (Meta AI)—is redefining this paradigm.
      </p>
      <p>
        Leveraging our co-founder's pioneering work in Reinforcement Learning
        with Human Feedback (RLHF) and Instruction-Following models at OpenAI,
        we've fundamentally redesigned LLMs for automation. By re-imagining the
        entire stack from the ground up, we optimize explicitly for
        AI-to-computer interactions rather than human-in-the-loop scenarios.
      </p>
      <p>Let's <b>automate the boring stuff</b>: joinus@typesafe.ai</p>
    </div>
  </body>
</html>