Yes, great idea to wear a surveillance device to replace the established tech known as âpencil and paper.â âMeta uses the example of running low on your favorite cereal. Your glasses will note when you inform them in the kitchen and remind them when youâre at the grocery store.â
Ray-Ban Meta glasses will receive live language translation abilities and video support later this year.
The perfect reading to understand the "Brian Sewell comes back as AI" story is the iconic Stochastic Parrots paper by @emilybender.bsky.social@mmitchell.bsky.socialdl.acm.org/doi/pdf/10.1...
![Text generated by an LM is not grounded in communicative intent, any model of the world, or any model of the reader's state of mind. It can't have been, because the training data never included sharing thoughts with a listener, nor does the machine have the ability to do that. This can seem counter-intuitive given the increasingly fluent qualities of automatically generated text, but we have to account for the fact that our perception of natural language text, regardless of how it was generated, is mediated by our own linguistic competence and our predisposition to interpret communicative acts as conveying coherent meaning and intent, whether or not they do [89, 140]. The problem is, if one side of the communication does not have meaning, then the comprehension of the implicit meaning is an illusion arising from our singular human understanding of language (independent of the model).22 Contrary](https://cdn.bsky.app/img/feed_fullsize/plain/did:plc:dvvvi2lfda3eq46zikkozitb/bafkreie3xjf67lsxaesenm3uho264tbxwevzr34gcvovj4ky6ghe4mrzte@jpeg)
![to how it may seem when we observe its output, an LM is a system for haphazardly stitching together sequences of linguistic forms it has observed in its vast training data, according to probabilistic information about how they combine, but without any reference to meaning: a stochastic parrot.
6.2 Risks and Harms
The ersatz fluency and coherence of LMs raises several risks, precisely because humans are prepared to interpret strings belonging to languages they speak as meaningful and corresponding to the communicative intent of some individual or group of individuals who have accountability for what is said. We now turn to examples, laying out the potential follow-on harms.
The first risks we consider are the risks that follow from the LMs absorbing the hegemonic worldview from their training data. When humans produce language, our utterances reflect our worldviews, including our biases [78, 79]. As people in positions of privilege with respect to a society's racism, misogyny, ableism,](https://cdn.bsky.app/img/feed_fullsize/plain/did:plc:dvvvi2lfda3eq46zikkozitb/bafkreicpijhznvdsvgo4h4w7kc367zuv7tk2nzd5in34vtszrxnnmxvvhq@jpeg)
'The London Evening Standard, rebranded as the London Standard, will feature a one-off AI-written review by the renowned critic, who died in 2015 at the age of 84, when it is published on Thursday.' Sometimes doing something simply because you can is not a good idea and not innovative.
âExperimental reviewâ to feature in relaunched newspaper âto provoke discussionâ about AI and journalism
Vice chancellors and provosts are suckers for a flashy presentation and words like âemergingâ âworld leadingâ âsynergisticâ part 90 gazillion
'In some cases, âinstitutional plagiarism checkers seem to be playing both sides of the marketâ, with some large edtech firms providing both a âpremium AIâŚto rephrase AI generated or normally plagiarised work so that it can avoid detectionâ and a plagiarism detector.' The squalor that is Edtech.
Essay mills pivoting to offering low-cost services to avoid plagiarism checks
Adding this to the ever-growing pile of evidence supporting the position that AI scraping isn't fair use. Because if you're willing to pay sometimes, you shouldn't get to just take the stuff you don't think is worth paying for.
