11 ms·

Anyone else find it weird that we haven't seen this from SpaceXAI/Grok?

• BryantD 6d ago
  Not really. They’re a couple of months behind OpenAI and Anthropic, although arguably ahead of the Chinese labs. And these attacks seem to require leading edge models. I worry that when and if Grok gets there, we’ll find out that SpaceXAI is too casual about security, though.
  • chrisjj 5d ago
    > And these attacks seem to require leading edge models. No. Just leading-edge irresponsibility.