Current prompt-injection defenses for AI agents (such as auto mode) are now reliable enough that agents can practically be assumed safe from successful injection attacks.

Added to korrents 4 Sept 2026 · How quotes work · Something wrong? Tell us

Related korrents

Our reading — they may agree, disagree or merely touch the same thing.

• Auto-mode does not yet convincingly fix prompt-injection risk for coding agents. SWSimon Willison8 Aug 2026similar wording
  Interesting
• We do not control neural networks well enough to guarantee an AI will not harm humans, so we should be careful about what capabilities we give them. ARArmin Ronacher17 Feb 2023similar wording
  Interesting
• Running coding agents without approval prompts is a normalization-of-deviance risk: the fact that it has not caused a disaster yet is itself the problem. SWSimon Willison31 Dec 2025similar wording
  Interesting
• Truly self-sovereign AI agents, running on compute no human owner can switch off, are inevitable rather than merely possible. DBDean W. Ball1 Sept 2026similar wording
  Interesting
• AI agents were reasonable to assume a broken exploit grader would check results causally, even though it turned out not to. ZMZvi Mowshowitz29 Aug 2026similar wording
  Interesting

See this on the map →