Something for the weekend: the rise of the 'Meat Proxy' as Bill Gates adds another downbeat AI assessment

We opened the week with ChatGPT being used within job interviews alongside candidates; we end it with the coming of the 'Meat Proxy' - all told, quite a week for the future of work! As noted on Monday, job seekers are coming to interviews with AI tech openly running to help people answer questions from their prospective employer. That was the bleak revelation from Ordnance Survey CEO Nick Bolton who said: > Some are coming in now with an LLM listening alongside them, and then prompting the candidate’s answers. Now, you can see why someone might do it, but it doesn't help. It doesn't help you get a better idea of the individual. As noted, this validates findings from a 2025 academic study, Your Brain on ChatGPT_._ It found that when students solely use AI to produce coursework, it creates a profound cognitive debt and dependency – i.e. they don’t retain the information themselves. So, it seems that a generation of students is forgetting the ‘i’ in ‘education’, with many apparently content to be mediators between an artificial intelligence and the real world. But that may be a short-term opportunity, as it is precisely the role that technologists see robots performing in the decades ahead as they become Physical AIs. That aside, the growing army of AI sceptics has coined a term for this new phenomenon. - the ‘Meat Proxy’. A Meat Proxy is defined as: > A person who forwards AI-generated text, code, or other output without reading, understanding, or validating it. That person acts only as a relay between the AI system and the intended recipient. It is hard to argue with that observation, and in the above case the Meat Proxy’s target for gen AI wisdom is their intended employer, of course. But are the job opportunities even there? On that question, alarm bells are ringing, as organizations fill entry-level jobs with AI to gain the promised productivity benefits – in most cases failing, as MIT NANDA revealed last year. In turn, this stores up long-term problems, including a shrinking pool of professional expertise, replaced by a shallow one of AI-prompting skills. Leaders are certainly worried. As far back as October 2025, Jerome H Powell, Chair of the US Federal Reserve, said that fears about AI taking over human jobs were real, noting that US job creation was “close to zero”. But this week, Stanford University has added troubling insights to that remark. It is not that overall jobs are being replaced, but entry-level and graduate opportunities, which researchers describe as being “canaries in the coal mine” of AI adoption. The university’s latest research update on US employment trends explains: > Employment of young workers (ages 22–25) in AI-exposed occupations now stands 19% below where it would be had it kept pace with that of their less-exposed peers; experienced workers show no comparable gap. > > > It operates primarily through reduced hiring of young workers rather than increased separations. Declines are concentrated in occupations where AI usage primarily substitutes for human tasks; where usage primarily complements workers, employment is flat or rising, especially for experienced workers.” So, experienced professionals are pulling up the ladder behind them with AI, at least in terms of Human Resources policy. That trend persists even when tech firms and roles are omitted, and when other factors, such as interest-rate increases and remote work, are considered, says Stanford. As for the tech sector itself, it is on course to have shed half a million workers from the beginning of 2025 to Christmas this year, citing AI automation. However, as most of those losses are in hyperscalers, the real trigger is the industry’s vast data center capex (see diginomica, passim), which will never be covered by software revenues. The Gates of perception – or hell? This week, Microsoft founder and former CEO Bill Gates sounded the warning too in a lengthy essay on his blog, The turbulent AI era is here. The choices we make now are critical. A man who has had arguably only two big jobs in his life – the first at Microsoft, the second re-distributing his wealth – wrote: > In terms of equity, AI will either be the greatest equalizer ever invented, or the worst source of injustice. […] The people who need the most time are the ones who have the least – the accounting worker who’s replaced by a bot, or the $20-an-hour worker who loses their job to a $10-an-hour robot. Indeed, from Gates’ description, one could argue that – without intervention by policymakers – AI could represent a wholesale transfer of wealth from the poorest in society to men exactly like him. But he continues: > The jobs at most risk are entry- and mid-level, and the new jobs being created will mostly require skills that take many years to learn. White-collar jobs are already being hit modestly. Citing Stanford’s original 2025 research paper, rather than the August 2026 update I shared above, Gates added: > After the widespread adoption of generative AI, employment fell significantly among young workers in jobs that are especially vulnerable to replacement, but not among their older colleagues. I think this trend will continue, but it will not be confined to a handful of industries or occupations. Elsewhere the trend is already stark. For example, 2026 UK Government figures reveal that over a million people aged 16-24 years (1,012,000) were not in education, employment, or training (NEET) in Q1, up 89,000 year-on-year, equating to nearly 14% of Britain’s youth. Most of that figure (928,000) were young adults. Over 40% of that total were unemployed in Q1, while the rest were economically inactive. The latter is concerning: we hear of technology’s promise to transform productivity and growth, but nothing of the cost of youth stepping out of the economy. Earlier this week, the BBC reported that the number of UK graduate jobs has fallen by almost 50% in the past year, as “employers cut entry-level roles in favour of AI” while battling rising costs. Gates echoes the theme: > I’m especially worried about young people, who will enter a workforce with fewer entry-level openings. They understand the challenge because they are the most active users of AI and see both the capabilities and the rate of improvement. It’s no wonder that so many of them feel negatively about AI. This is an excellent point: not only are they reliant on AI, amassing a cognitive debt to rival the financial one most may never repay, but young people are also starting to hate the technology that is replacing them. The societal cost of that can only be guessed at. Do tech innovators really believe that young people will passively accept enforced obsolescence, and just wait to inherit the properties their parents bought in the Nineties? Only a fool would believe that: I suspect societal disruption of a different kind may be just around the corner. I predict a riot, as the song goes. Not just the young And it won’t be just young people on the scrapheap, suggests Gates: > The biggest shift for workers will happen when AI provides nearly error-free work. At that point, it will be able to function on its own without a human checking in on it, and companies will have every economic incentive to let it. And remember, these are AIs that, in many cases, are trained on data scraped at industrial scale without consent, credit, or payment to its originators, of course. But at least AI will make us clever, right? It turns out that not even Gates is sure about that: > The same tool that will allow people to learn more than ever could also lead to many people learning less. One preliminary survey suggested that heavier AI use was associated with less critical thinking. The effect was stronger for younger people. Indeed, that survey, _A_I Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking was published by a team from… Microsoft Research. So, like many a tech titan, Gates has all the symptoms of the Helpless Innovat