Interpreting is the oral version of translation. Currently, almost all interpreting is done by humans, although people use the spoken-word features of Google Translate and similar services while traveling, and services like OpenAI's Whisper are working on speech recognition/translation combinations. Professional services like Skype Translation seem not to have gained popularity.
The market will resolve Yes once three separate meetings hosted by the US President (=involving the US President and held in the US) have been interpreted by a computer into three different languages.
See also:
Update 2025-09-01 (PST) (AI summary of creator comment): - The interpreting that counts must be the official interpretation used in the meeting—i.e., the interpretation between the US President and their guest must be performed by a computer (not a human).
1,000People are also trading
Under some definitions, they are already unnecessary. Under others, approximately never until we get to AGI. This (and the two linked one) are very much "know thy resolution criteria"... and these criteria are a shocking combination of specific and vague. Who is supposed to interpret? What are requirements for quality? Like, wha...?
@b575 what exactly do you mean? I assume there can be edge cases, but I tried to make the resolution criteria as objective as possible...
@PS Reread the question - I meant that the official interpretation between the US president and his guest will be performed by a computer, not a human. Was that the issue?
@PS Part of the issue, but also, does it not matter what the quality is and does it not matter what the social pressures on the matter are?
@b575 I assume that the quality will have to be good enough if it to be used at such a level, and multiple times at that. The question about social pressures I don't understand.
@PS "Good enough" is hella vague. And stakes may differ enormously.
The social pressures part is pretty straightforward: the main barrier before politicians talking through translation software is the prestige of having a human translator (and some off-the-cuff lubrication said translator can provide for the more emotional comments), not the actual danger of misunderstanding.
@b575 I still don't really understand your criticism. I don't claim my resolution criteria have an ideal or even exact correspondence to the question of when human interpreters will become unnecessary, and it was the most useful objective criterion I came up with. If you have a better one, I'd be interested in seeing that market and happy to boost it.
@PS I agree that this is a difficult thing to operationalize, but I also don't think I would come up with any operationalization that wouldn't resolve YES on creation. What are you trying to capture?