3 ms·
As someone who just got out of a meeting demonstrating how Copilot running Luna can be breadcrumbed by a single line of text in innocuous package into downloadi
by ofjcihen 2mo ago
As someone who just got out of a meeting demonstrating how Copilot running Luna can be breadcrumbed by a single line of text in innocuous package into downloading and installing malware I think Boris Cherny might be wrong.
- keeda 2mo agoBoris is talking about Anthropic models only of course. In the chart in his tweet Luna has like a 43% chance of being compromised with sufficient attempts. But even the Anthropic models have as high as a 5% chance, which is still far too much IMO.
- ofjcihen 2mo agoThat’s fair and to their credit (and somewhat the user’s detriment) I think the user focused guardrails they’ve imposed have unintentionally helped to protect against the kind of attacks I have success with. But you’re right…5 percent is better than what it was but at scale that’s still quite a bit.