Why the New York Times sent this chatbot creator a cease and desist

It’s probably not a good idea to use the word disintermediating on a Saturday.

Yet the legal term might be one you should learn if you are developing a new chatbot or already have one available. In a recent post, Chris McBride — the creator of the Julia cooking bot — explained how he was contacted by the legal team at the New York Times telling him to stop using their recipes. It turns out he wasn’t really “using” them, though. In a demo, you can see that the bot mentions the recipe and who made it, then links to the Times. Many chatbots share links, even if it’s to another article or a restaurant chain. It sure doesn’t seem like a heinous violation of anything, but the infringement lacks disintermediation, that is — a go-between that explains why the links are presented.

[aditude-amp id="flyingcarpet" targeting='{"env":"staging","page_type":"article","post_id":2034506,"post_type":"story","post_chan":"none","tags":null,"ai":false,"category":"none","all_categories":"bots,","session":"A"}']

A chatbot don’t always show which links are used or how many it contains. On a website like VentureBeat, you can easily see that this is a link to an article about chatbots, and you can even look at the code on this page to inspect the link and see how it works. Chatbots present links within a chat, so you can’t inspect the code, and you don’t know how many links are contained in the bot.

As McBride points out, there is a legal precedence for linking to other sites — and it is acceptable use — but there isn’t much of a precedence for chatbots doing that. He calls for a linking policy. The bot development community should probably come up with a more obvious way to show that a link in a chat takes you to a website and is not something the bot came up with on its own.

You can read the full Medium post here.