\n<\/p>\n
In Silicon Valley, some of the brightest minds believe a universal basic income (UBI) that guarantees people unrestricted cash payments will help them to survive and thrive as advanced technologies eliminate more careers as we know them, from white collar and creative jobs \u2014 lawyers, journalists, artists, software engineers \u2014 to labor roles<\/a>. The idea has gained enough traction that dozens<\/a> of guaranteed income programs have been started in U.S. cities since 2020.<\/p>\n Yet even Sam Altman, the CEO of OpenAI and one of the highest-profile proponents<\/a> of UBI, doesn\u2019t believe that it\u2019s a complete solution. As he said during a sit-down<\/a> earlier this year, \u201cI think it is a little part of the solution. I think it\u2019s great. I think as [advanced artificial intelligence] participates more and more in the economy, we should distribute wealth and resources much more than we have and that will be important over time. But I don\u2019t think that\u2019s going to solve the problem. I don\u2019t think that\u2019s going to give people meaning, I don\u2019t think it means people are going to entirely stop trying to create and do new things and whatever else. So I would consider it an enabling technology, but not a plan for society.\u201d<\/p>\n The question begged is what a plan for society should then look like, and computer scientist Jaron Lanier, a founder in the field of virtual reality, writes in this week\u2019s New Yorker<\/a> that \u201cdata dignity\u201d could be one solution, if not the<\/em> answer.<\/p>\n Here\u2019s the basic premise: Right now, we mostly give our data for free in exchange for free services. Lanier argues that in the age of AI, must we stop doing this, that the powerful models currently working their way into society \u201cbe connected with the humans\u201d who give them so much to ingest and learn from in the first place.<\/p>\n The idea is for people to \u201cget paid for what they create, even when it is filtered and recombined\u201d into something that\u2019s unrecognizable.<\/p>\n The concept isn\u2019t brand new, with Lanier first introducing the notion of data dignity in a 2018 Harvard Business Review piece titled, \u201cA Blueprint for a Better Digital Society<\/a>.\u201d<\/p>\n As he wrote at the time with co-author and economist Glen Weyl, \u201c[R]hetoric from the tech sector suggests a coming wave of underemployment due to artificial intelligence (AI) and automation.\u201d But the predictions of UBI advocates \u201cleave room for only two outcomes,\u201d and they\u2019re extreme, Lanier and Weyl observed. \u201cEither there will be mass poverty despite technological advances, or much wealth will have to be taken under central, national control through a social wealth fund to provide citizens a universal basic income.\u201d<\/p>\n The problem is that both \u201chyper-concentrate power and undermine or ignore the value of data creators,\u201d they wrote.<\/p>\n Of course, assigning people the right amount of credit for their countless contributions to everything that exists online is not a minor challenge. Lanier acknowledges that even data-dignity researchers can\u2019t agree on how to disentangle everything that AI models have absorbed or how detailed an accounting should be attempted.<\/p>\n Still, he thinks that it could be done \u2014 gradually.\u00a0 \u201cThe system wouldn\u2019t necessarily account for the billions of people who have made ambient contributions to big models\u2014those who have added to a model\u2019s simulated competence with grammar, for example.\u201d But starting with a \u201csmall number of special contributors,\u201d over time, \u201cmore people might be included\u201d and \u201cstart to play a role.\u201d<\/p>\n Alas, even if there is a will, a more immediate challenge \u2014 lack of access \u2014 looms. Though OpenAI had released some of its training data in previous years, it has since closed the kimono completely. When Greg Brockman described to TechCrunch last month<\/a> the training data for OpenAI\u2019s latest and most powerful large language model, GPT-4,\u201d he said it derived from a \u201cvariety of licensed, created, and publicly available data sources, which may include publicly available personal information,\u201d but he declined to offer anything more specific.<\/p>\n As OpenAI stated<\/a> upon GPT-4\u2019s release, there is too much downside for the outfit in revealing more than it does. \u201cGiven both the competitive landscape and the safety implications of large-scale models like GPT-4, this report contains no further details about the architecture (including model size), hardware, training compute, dataset construction, training method, or similar.\u201d (The same is true of every large language model currently, including Google\u2019s Bard chatbot.)<\/p>\n Unsurprisingly, regulators are grappling with what to do. OpenAI \u2014 whose technology in particular is spreading like wildfire \u2014 is already in the crosshairs of a growing number of countries, including the Italian authority, which has blocked the use of its popular ChatGPT chatbot. French, German, Irish, and Canadian data regulators are also investigating how it collects and uses data.<\/p>\n But as Margaret Mitchell, an AI researcher who was formerly Google\u2019s AI ethics co-lead, tells the outlet\u00a0 Technology Review<\/a>, it might be nearly impossible at this point for these companies to identify individuals\u2019 data and remove it from their models.<\/p>\n As explained by the outlet: OpenAI would be better off today if it had built in data record-keeping from the start, but it\u2019s standard in the AI industry to build data sets for AI models by scraping the web indiscriminately and then outsourcing some of the clean-up of that data.<\/p>\n If these players truly have a limited understanding of what\u2019s now in their models, that\u2019s a pretty big challenge to the \u201cdata dignity\u201d proposal of Lanier, who calls Altman a \u201ccolleague and friend\u201d in his New Yorker piece.<\/p>\n Whether it renders it impossible is something only time will tell.<\/p>\n Certainly, there is merit in determining a way to give people ownership over their work, even if it\u2019s made outwardly \u201cother.\u201d It\u2019s also highly likely that frustration over who owns what will grow as more of the world is reshaped with these new tools.<\/p>\n Already, OpenAI and others are facing numerous<\/a> and wide-ranging copyright infringement lawsuits<\/a> over whether or not they have the right to scrape the entire internet to feed their algorithms.<\/p>\n Perhaps even more importantly, giving people credit for what comes out of these AI systems could go help preserve humans\u2019 sanity over time, suggests Lanier in his fascinating New Yorker piece.<\/p>\n People need agency, and as he sees it, universal basic income alone \u201camounts to putting everyone on the dole in order to preserve the idea of black-box artificial intelligence.\u201d<\/p>\n Meanwhile, ending the \u201cblack box nature of our current AI models\u201d would make an accounting of people\u2019s contributions easier \u2014 which might make them far more likely to continue making contributions.<\/p>\n It might all boil down to establishing a new creative class instead of a new dependent class, he writes. And which would you prefer to be a part of?<\/p>\n<\/p><\/div>\nUntangle my mind<\/h2>\n
How to save a life<\/h2>\n