This webinar is best for:
- Business decision makers
- Security officers
- Technology professionals
AI agents can be powerful, but their use necessitates the creation of accountability mechanisms and collaboration frameworks.
Trust in AI systems is crucial, and it is important to design AI systems in a way that ensures data security, content safety, and transparency. As AI agents become increasingly prevalent in our lives and workplaces, there is a need for designing AI agents that are accountable, trustworthy, and collaborative. Establishing guardrails to control the actions of AI agents is essential to prevent them from overreaching or executing actions they have not been allowed to perform.
Originally presented at Slack City Tour London, join this webinar to hear an unbiased view from Franny Hsiao, Distinguished AI Architect at Salesforce.
Featured speakers:
