[scrupulusalbion]

I doubt this could ever be used to arbitrarily read minds. Even if you could do the measurements without the targer knowing, you would have to somehow correlate the target's real thoughts with those measurements. In the article, the targets voluntarily provide that correlation to measurements during a training phase. It might be the case that some neuro-physical patterns might be common across targets, but that seems very unlikely.

If the target is in your custody and you coerce him to give training data, then he could just give garbage data. You tell him to think "fish barrel neuton whale", but he really thinks "somebody once told me".

Do sub-vocalizations only occur for certain categories of thoughts? I suspect that facial memories do not induce sub-vocalizations.

[schoen]

Maybe a more intrusive setting would be something like a polygraph. The subject would be first tested to ensure that the training worked to a high accuracy level (like the polygraph baseline measurement). You could tell to some extent whether the subject was cooperating with this training because you could see the resulting accuracy rates, and maybe continue the training until they improved, or else declare the subject to have failed to cooperate with the test.

Probably, as with an present-day polygraph, people could specifically train themselves to produce inaccurate results, but also presumably most people wouldn't have done so.

On the other hand, I don't know what level of language this system operates at so I don't know if you're detecting phonemes, words, phoneme features (like +voice/-voice), or what. This seems significant for understanding how usable it would be for "mind reading" if the subject's thoughts included many words that hadn't been individually trained. If you're just detecting conditions that you've already trained for, it doesn't seem like there's really anything new here (from the mind-reading point of view, as opposed to the UI point of view) compared to existing biofeedback or polygraph systems because you're mainly just detecting when a known physiological state is or isn't re-elicited, which is something the earlier systems could already do.

Edit: other commenters said they can only distinguish about 20 words, so presumably the technology has a long way to go before it would be relevant to semi-adversarial mind reading.

[zorkw4rg]

how can you possibly claim this so confidently, but sure keep believing its impossible, until its not, but then its too late to worry about anyways. This seems like an incredibly foolish opinion on this matter.

[Roboprog]

If nothing else, I suppose you could also train yourself to think in code.

That word you keep using - I don’t think it means what I think you think it means.

That sort of thing.