{"id":1341,"date":"2026-11-18T12:14:08","date_gmt":"2026-11-18T12:14:08","guid":{"rendered":"https:\/\/snap-trove.co.za\/?p=1341"},"modified":"2026-11-18T12:14:08","modified_gmt":"2026-11-18T12:14:08","slug":"can-ai-read-your-mind-how-close-are-we-really","status":"publish","type":"post","link":"https:\/\/snap-trove.co.za\/index.php\/2026\/11\/18\/can-ai-read-your-mind-how-close-are-we-really\/","title":{"rendered":"Can AI Read Your Mind? How Close Are We Really?"},"content":{"rendered":"<p><span style=\"font-weight: 400;\">For most of human history, reading another person&#8217;s thoughts belonged to the realm of magic and science fiction. In 2026, machines can already do something that looks remarkably similar: record brain activity and turn it into language, sometimes in real time.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In June, Meta unveiled Brain2Qwerty v2, an artificial intelligence system that reconstructs sentences from brain activity without putting anything inside the skull. Then, in September, researchers at the University of California, San Francisco, demonstrated a brain implant capable of decoding speech and gestures simultaneously, allowing people with severe paralysis to communicate through a digital avatar.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The technology is still experimental. But the direction is unmistakable: AI is becoming increasingly capable of translating the activity of the brain into something a computer can understand.<\/span><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-1343\" src=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/AI-reading-mind-300x198.jpg\" alt=\"\" width=\"300\" height=\"198\" srcset=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/AI-reading-mind-300x198.jpg 300w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/AI-reading-mind-1024x675.jpg 1024w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/AI-reading-mind-768x507.jpg 768w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/AI-reading-mind.jpg 1093w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/p>\n<p><!--nextpage--><\/p>\n<h2><span style=\"font-weight: 400;\">An AI That Turns Brain Activity Into Sentences<\/span><\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-1344\" src=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/brain2qwerty-300x198.jpg\" alt=\"\" width=\"300\" height=\"198\" srcset=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/brain2qwerty-300x198.jpg 300w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/brain2qwerty-1024x675.jpg 1024w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/brain2qwerty-768x507.jpg 768w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/brain2qwerty.jpg 1093w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Meta&#8217;s Brain2Qwerty v2 works with magnetoencephalography, or MEG, a technique that detects the extremely weak magnetic fields generated by electrical activity in the brain. Unlike an implanted brain-computer interface, MEG requires no surgery. <\/span><span style=\"font-weight: 400;\">Researchers trained the system using recordings from nine healthy volunteers. Each participant spent around 10 hours in the MEG scanner while typing, producing roughly 22,000 sentences for the study. The AI learned to associate patterns of brain activity with letters, words and entire sentences.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The results were striking. Brain2Qwerty v2 achieved an average word accuracy of 61%, compared with around 8% for previous non-invasive approaches. The best-performing participant reached 78% word accuracy, with more than half of the decoded sentences containing one word error or fewer. The previous Brain2Qwerty version reached 48% in its best participant.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The researchers found another important pattern: performance continued to improve as more training data was added. So far, they have not found a clear point where the improvement stops. That raises the possibility that larger datasets and more powerful AI models could push brain-to-text systems considerably further.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The crucial detail is what the system is actually decoding. Brain2Qwerty is not scanning someone&#8217;s mind and producing a transcript of every thought. It is learning the neural patterns associated with a highly specific task: producing typed sentences. The breakthrough lies in how effectively AI can extract linguistic information from an extremely noisy biological signal.<\/span><\/p>\n<p><!--nextpage--><\/p>\n<h2><span style=\"font-weight: 400;\">Giving a Voice\u2014and a Body\u2014Back to Paralysed Patients<\/span><\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-1345\" src=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/voice-to-paralized-patients-300x198.jpg\" alt=\"\" width=\"300\" height=\"198\" srcset=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/voice-to-paralized-patients-300x198.jpg 300w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/voice-to-paralized-patients-1024x675.jpg 1024w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/voice-to-paralized-patients-768x507.jpg 768w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/voice-to-paralized-patients.jpg 1093w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/p>\n<p><span style=\"font-weight: 400;\">On September 14, researchers at UCSF reported a brain-computer interface that can decode speech and physical gestures at the same time. The system uses a high-density electrocorticography implant containing 253 electrodes placed directly on the surface of the brain.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The researchers tested the technology in people with severe paralysis, including individuals affected by brainstem stroke and ALS. The participants attempted to speak or perform gestures such as waving or giving a thumbs-up, even when their bodies could no longer physically carry out those movements. <\/span><span style=\"font-weight: 400;\">AI translated the resulting brain signals into commands for a personalised full-body virtual avatar. Speech appeared as text, while gestures could make the avatar move. The system could even decode the two forms of communication simultaneously, allowing a participant to conduct a conversation using words and body language together.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">That distinction matters because human communication has never been purely verbal. A nod, shrug, wave or facial expression can completely change the meaning of a sentence. For someone who has lost the ability to move or speak, restoring those signals could make communication dramatically more natural. <\/span><span style=\"font-weight: 400;\">In one conversational test, a participant achieved median decoding accuracies of 100% for both speech and gestures across three blocks. The researchers describe the work as a proof of concept for a unified communication system rather than a finished clinical device.<\/span><\/p>\n<p><!--nextpage--><\/p>\n<h2><span style=\"font-weight: 400;\">Where Does Mind Reading Begin?<\/span><\/h2>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"alignnone size-medium wp-image-1342\" src=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/ai-mind-reading-300x198.jpg\" alt=\"\" width=\"300\" height=\"198\" srcset=\"https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/ai-mind-reading-300x198.jpg 300w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/ai-mind-reading-1024x675.jpg 1024w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/ai-mind-reading-768x507.jpg 768w, https:\/\/snap-trove.co.za\/wp-content\/uploads\/2026\/11\/ai-mind-reading.jpg 1093w\" sizes=\"auto, (max-width: 300px) 100vw, 300px\" \/><\/p>\n<p><span style=\"font-weight: 400;\">Together, these experiments expose a fascinating boundary. <\/span><span style=\"font-weight: 400;\">AI can already extract meaningful information from brain activity. It can recognise patterns associated with intended language, reconstruct words and sentences, and identify attempted physical movements. The machine does not need to understand consciousness in some mysterious, human sense. It needs enough data to learn the relationship between neural activity and behaviour.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">For people trapped inside bodies they cannot control, brain-computer interfaces could eventually provide communication that is faster, richer and more expressive than conventional eye-tracking systems. A person who cannot speak or move could communicate through words, gestures and expressions generated directly from neural activity.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">The same technology however creates an entirely new category of privacy problem.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>For most of human history, reading another person&#8217;s thoughts belonged to the realm of magic and science fiction. In 2026, machines can already do something that looks remarkably similar: record&#8230;<\/p>\n","protected":false},"author":1,"featured_media":1343,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2,6],"tags":[],"class_list":["post-1341","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-exclusives","category-science"],"_links":{"self":[{"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/posts\/1341","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/comments?post=1341"}],"version-history":[{"count":1,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/posts\/1341\/revisions"}],"predecessor-version":[{"id":1346,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/posts\/1341\/revisions\/1346"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/media\/1343"}],"wp:attachment":[{"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/media?parent=1341"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/categories?post=1341"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/snap-trove.co.za\/index.php\/wp-json\/wp\/v2\/tags?post=1341"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}