Tag
OpenAI publishes guidance on designing AI agents resistant to prompt injection attacks, arguing that modern attacks increasingly use social engineering tactics rather than simple string injections, and advocating for system-level defenses that constrain impact rather than relying solely on input filtering.
Anthropic's Vend experiment showcases how the AI agent Claudius end-to-end managed a store in an office vending machine, revealing challenges such as social engineering attacks and solutions for profitability through a multi-agent architecture.