Can a machine fool you into thinking it's a live human being? CyberLover.ru will sell an IM bot in February, intended to seduce the lovelorn. Connor Sweeney's Reuters report says this technology can be used for evil purposes.

Can you prove you are a human? From within a Skype chat?

Or test that everyone on the other end of the chat is human too?

Should Skype offer this credentialing, making it universal? Or should it come from the anti-virus community?

How can we certify that robots in a chatroom are trusted? Have not been hacked? Follow the chat room's privacy rules?

What are well-understood visual indicators for these four levels of in-chat trust and authenticity?

• A confirmed human
• A trusted robot
• A robot
• Unknown

CRM and IVR systems use robots for screening, routing and after-hours customer service. How well do you disclose your robots' botness? How cleanly do you handle the transfer from a bot to a human or from a human to a bot? What are your organization's formal guidelines?