{"id":258255,"date":"2026-09-28T09:11:27","date_gmt":"2026-09-28T06:11:27","guid":{"rendered":"https:\/\/digitrendz.blog\/z\/?p=258255"},"modified":"2026-09-28T09:11:27","modified_gmt":"2026-09-28T06:11:27","slug":"engram-turns-broken-ai-hallucinations-into-music","status":"publish","type":"post","link":"https:\/\/digitrendz.blog\/z\/tech-news\/258255\/engram-turns-broken-ai-hallucinations-into-music\/","title":{"rendered":"Engram Turns Broken AI Hallucinations Into Music"},"content":{"rendered":"
\u25bc Summary<\/summary>

– Music startup Thoughtful Things has launched a Kickstarter campaign for Engram, its first AI-powered sampler and groovebox.
– The device utilizes locally running ‘tiny AI’ models to mangle audio and hallucinate new sounds without requiring an internet connection.
– The company emphasizes ethical training by using only commercially licensed open datasets and plans to release open firmware for user customization.
– Founder Evan King describes the instrument as a field recorder for latent space, encouraging experimental sound design inspired by circuit bending.
– Early pledges start at $675 with estimated retail prices between $850 and $900 following the conclusion of the limited campaign run.
<\/p><\/details>\n\n

<\/p>\n\n

T<\/mark>houghtful Things<\/strong> has officially launched the Kickstarter<\/a> campaign<\/strong> for its debut hardware instrument, Engram<\/a><\/strong>. This device functions as a sampler and groovebox that leverages artificial intelligence to distort incoming audio streams and generate entirely new sonic textures. The product is not designed to be a consumer-friendly song generator akin to Suno<\/a>, nor is it intended to produce radio-ready pop tracks. Instead, Engram targets experimental musicians seeking to push AI audio models to their breaking points, creating uncanny and glitch-heavy soundscapes.<\/p>\n\n

A key differentiator for the device is its offline operation. Engram does not rely on cloud connectivity; instead, it processes audio using a locally hosted “tiny AI.” The company emphasizes that this proprietary model was custom-trained in-house using only open datasets comprising audio licensed for commercial use, such as those under CC-BY licenses.<\/p>\n\n

\u201cWe\u2019ve trained our audio models on open datasets that only contain audio licensed for commercial use (CC-BY or similar). We have not trained and will never train our models on non-commercial, pirated, or otherwise stolen data.\u201d<\/p>\n\n

This commitment to ethical training data is paired with a philosophy of creative disruption. Founder Evan King<\/a><\/strong> describes the Engram as a “field recorder for latent space,”<\/strong> inviting users to explore the unpredictable edges of AI generation. The hardware draws inspiration from circuit bending, allowing performers to manipulate and intentionally break the underlying models to achieve unique results. In promotional demonstrations, King uses voice commands to request specific instruments, resulting in distorted outputs that retain only a vague resemblance to the source material.<\/p>\n\n

To foster further innovation, Thoughtful Things<\/a> plans to release the Engram\u2019s firmware, enabling developers and musicians to tweak existing parameters or load custom AI models. The initial production run is exclusive to the crowdfunding campaign, with early bird pledges starting at $675<\/strong>. While a standard retail price has not been confirmed, the current discount represents a 30 percent reduction, suggesting a future market price likely falling between $850 and $900<\/strong>.<\/p>\n\n(Source: The Verge<\/a>)<\/em>","protected":false},"excerpt":{"rendered":"

Thoughtful Things has launched a Kickstarter campaign for Engram, an offline AI-powered sampler and groovebox designed for experimental musicians to create glitch-heavy soundscapes. The device utilizes a locally hosted model trained exclusively on commercially licensed open datasets, ensuring eth…<\/p>\n","protected":false},"author":1,"featured_media":258254,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_themeisle_gutenberg_block_has_review":false,"cybocfi_hide_featured_image":"","footnotes":""},"categories":[57,3247,6451,3327,3254],"tags":[258843,258844,257102,203176,79327],"entities":[258845,258847,17001,9424,2742,258846],"class_list":["post-258255","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tech-news","category-artificial-intelligence","category-gadgets","category-newswire","category-technology","tag-engram","tag-evan-king","tag-kickstarter","tag-suno","tag-the-verge","entity-engram","entity-evan-king","entity-kickstarter","entity-suno","entity-the-verge","entity-thoughtful-things"],"_links":{"self":[{"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/posts\/258255","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/comments?post=258255"}],"version-history":[{"count":2,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/posts\/258255\/revisions"}],"predecessor-version":[{"id":258272,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/posts\/258255\/revisions\/258272"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/media\/258254"}],"wp:attachment":[{"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/media?parent=258255"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/categories?post=258255"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/tags?post=258255"},{"taxonomy":"entity","embeddable":true,"href":"https:\/\/digitrendz.blog\/z\/wp-json\/wp\/v2\/entities?post=258255"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}