No Ifs, Ands, or Bots! Bot Use Disclosure Is Mandatory in California
As anyone who shops or interacts online knows, many product and service providers offer messaging options that allow visitors to ask questions and interact with the site. Often, the “person” they’re communicating with isn’t real, although it often seems human – “Hi! I’m Tina! Can I help you?”
It’s a chatbot, a type of bot that simulates conversations online. In the last few years, artificial intelligence has made these types of bots less rigid in their responses, and more intuitive, convincing, and persuasive. (One example is that Sephora has bots that provide makeup tutorials).
So, why should you care? You should care if bots are involved in your offering goods or services online. Or, if you’re an influencer or blogger who uses bots to push content out on social or otherwise interact with followers.
California has a bot disclosure (now referred to in this blog as the “Law”) that requires clear and conspicuous disclosures when bots are used to communicate or interact online with people in California. It’s doesn’t matter whether the bot provider is based in California or not.
The Law says that’s it’s unlawful for any person to use a bot to communicate or interact online with someone in California “with the intent to deceive the person about its artificial identity or for the purpose of knowingly deceiving the person (“knowingly” often means not knowing about, as well as ignoring the law) about the content of the communication”, for purposes of soliciting or incentivizing a commercial transaction.
Examples of soliciting or incentivizing aren’t just getting someone to purchase or sell goods or services. It can also be such activities as suggesting a credit or alternate item instead of a refund, or helping a visitor find a particular type of restaurant). Violating these laws could lead to a reprimand or fine of up to $2,500 per violation from the California Attorney General, whose office constantly monitors online for violators.
(That fine could be expensive. A “violation” is a bot speaking to one person. So, if your bot speaks to a hundred people and you get caught, it could cost you $250,000 (!).
So, when in doubt, disclose.
Unfortunately, the Law doesn’t say how bot disclosures should be made. But, based on guidance in the Federal Trade Commission Act (the same law that applies to influences disclosing endorsements), I think a best practice and one that should comply with the Law is to conspicuously and clearly state that a bot you’re using is a bot. And the disclosures should be repetitive. Here’s some examples:
The initial popup chat box should say something like:
You should repeat the disclosure if the bot’s engagement with the visitor is lengthy. A bot’s response to a visitor’s message after a certain number of inquiries could be
A simpler disclaimer like “Hi! I’m Bob! I’m not a person” may not comply.
So, avoid problems with the Law. Say if it’s a bot or not.
(By the way, I’m not a bot!)
(C) 2021 Paul I. Menes