{"text":[[{"start":7.2,"text":"On Air Force One last week, US President Donald Trump caught many off guard by throwing his weight behind a proposal that would see the US take equity stakes in AI companies. Ensuring society shares in the gains from the technology is an admirable goal, and seizes on a growing anti-AI backlash in America. But rather than democratising the wealth created by new technologies, the still fuzzy plan to seed an AI wealth fund with voluntary contributions from companies only risks strengthening the power of tech giants — and government administrators. "}],[{"start":41.550000000000004,"text":"Trump said he would discuss the initiative with AI companies this week, as he tries to boost his popularity on a topic that is gaining political traction on both the left and the right. Indeed, in recent weeks, several proposals have emerged."}],[{"start":57.35000000000001,"text":"Last week, Senator Bernie Sanders met with OpenAI’s Sam Altman, and laid out his argument for creating a sovereign wealth fund with “a one-time 50 per cent tax not on profits of OpenAI, Anthropic, xAI and other companies, but paid with something far more valuable than that: the stock”. What he describes sounds less like taxation, and more like nationalisation. Even if his plan were to pass Congress, it is likely to be challenged by the courts — the Fifth Amendment to the US constitution limits the public acquisition of private property without just compensation."}],[{"start":92,"text":"The government could buy shares in AI companies and compensate them at market rates. But in a country that is running large fiscal deficits, funding a sovereign wealth fund with debt issuance would risk swelling an already bloated national debt. It also risks crowding out private sector investment in AI companies, in part, by pushing up US bond yields, and it adds risk to the federal balance sheet."}],[{"start":115.75,"text":"The most likely path, then, is the one currently being considered by Trump: a fund that solicits voluntary equity contributions from technology companies. But why would AI firms choose to make voluntary contributions? Contributing to such a fund would curry favour with an administration that has shown its eagerness to use its powers to help friends and harm foes. Trump could give considerable concessions to those that contribute to the scheme — or bully those who refuse to chip in. Either way, the risk of political favouritism hampers the efficiency and governance of the AI sector. "}],[{"start":152.95,"text":"Putting shares in every American’s portfolio would give AI firms political cover, too. The federal government’s imprimatur risks creating pressure for bailouts in a downturn. If AI stocks continue to soar, Trump would be hailed a hero and every American would gain. But if they sink, taxpayers could be stuck with the bill. Giving every American a stake in just a few publicly traded equities is a recipe for moral hazard."}],[{"start":179.5,"text":"The voluntary contribution plan would create myriad conflicts of interest and primarily benefit its own architects. It also overlooks that many Americans are already benefiting from faster growth, productivity gains and higher equity valuations as a result of AI. The still vague current proposals to spread the gains of AI companies risk crimping the development of the nascent technology itself. For now it may be best to prioritise growth, alongside sensible regulatory guardrails that would ensure the technology is developed safely. In time, tax reforms to ensure the gains are more broadly shared might then be necessary. "}],[{"start":217.85,"text":"For years politicians have taken for granted that government should not pick winners and losers. Trump’s talks with AI companies should not change that."}],[{"start":233.7,"text":""}]],"url":"https://audio.ftcn.net.cn/album/a_1781139029_5766.mp3"}