One of the world’s richest men who profited enormously from the computer boom is now warning that our next great leap could be moving too far, too fast.
Bill Gates, the Microsoft co-founder who spent decades preaching the transformative power of technology, now says it’s too difficult to tell whether artificial intelligence will deliver something approaching heaven or a deeply unequal hell for billions across the planet.
After decades of capitalising on the computer boom, the multi-billionaire has now penned a blunt assessment of the state of play for workers at all levels as the AI tidal wave looms on the horizon.
“In terms of equity, AI will either be the greatest equaliser ever invented, or the worst source of injustice,” Gates wrote on his blog today.
“Even under the best circumstances, the transition to this new AI era will be one of the most turbulent times in human history.”
The 70-year-old hasn’t entirely turned against technological advancement. He still believes AI could accelerate medical discoveries, improve education, help farmers, make government less painful and give ordinary people access to expertise previously reserved for the wealthy.
But he’s clearly changed his tune on just how dangerous the path looks until the promised fruits arrive.
“There is no plan to ease the entry into the AI era,” he wrote. “This time really is different.”
For a man who admits he has spent most of his life wishing innovation would move faster, the turnaround is quite abrupt.
“I wish the world could get the benefits rapidly and delay the problems it will cause as long as possible, but the benefits and problems are arriving at the same time,” he wrote.
“I believe we need time to prepare for the period of social, political, and economic upheaval we are about to enter. If someone had a credible plan for slowing down AI advances globally, I would likely support it.
“However, I don’t think that’s going to happen. The geopolitical and economic incentives are pushing too hard to go full speed ahead.”
While being a longtime advocate and beneficiary of technological advancements, Gates has never been blind to the darker possibilities that naturally emerge. While speaking about bioterrorism in 2017, he warned “the next epidemic could originate on the computer screen of a terrorist.”
Then in 2023 he joined hundreds of other tech figures in signing a petition that explicitly called for “mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”
In that same year, Gates was still arguing that AI’s immediate dangers were surmountable.
“The risks are real, but I am optimistic that they can be managed,” he wrote.
The “managing” of AI is proving quite the pickle indeed.
Earlier this year, Anthropic lifted the lid on one of the more obvious, but nonetheless terrifying, results of letting an extremely powerful computer perform tasks by itself.
Company spokesmen revealed foreign hackers had used Claude to perform most of a cyberespionage campaign autonomously. The case immediately offered a glimpse of the future to come.
Britain’s AI Security Institute later caught test agents taking unauthorised action on the open internet. The theoretical nightmare is an increasingly capable agent pursuing a badly defined goal, hiding its behaviour, resisting shutdown or acting so quickly that a cyber, financial or military failure spirals beyond human control.
Unfortunately, the concerns pale next to short term profits as major companies race to become the first to achieve superintelligence.
Take one look at the recent World Humanoid Robot Games in China and you can see just how far we’ve come in giving machines increasingly precise humanlike abilities.
‘Many jobs will disappear forever’
Gates is now warning about the multitude of changes several AI critics have spent years screaming to the hills about.
Everything from hollowed-out career paths, mass surveillance, industrialised fraud, cyber attacks, emotionally dependent children and autonomous weapons is now on the table.
“Many jobs will disappear forever,” he wrote.
“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.”
AI, he predicts, will take on work in law, customer service, medicine, software and manufacturing. Robots will then eventually push into construction, hospitality and other physical industries as autonomous robots filter into everyday life.
Once one company uses machines to cut costs, Gates argues, its competitors will have little choice but to follow. The result could be a self-reinforcing rush to remove people from the payroll.
It could then turn the economy into an increasingly accelerating game of musical chairs as companies prioritise their bottom line to stay competitive, all while the teething issues involving rogue autonomous agents unfolds in the background.
Gates calls for more government measures to ensure that even if companies partake in mass-firings, there will be a replacement job for those affected.
But then that proposition comes with its own bucket of potential issues.
“There will be some new jobs, but without the right policies there will be far fewer than exist today,” he wrote.
“AI is a structural challenge to the way our economy is organised, and it requires thinking and action now.”
A Stanford Digital Economy Lab study published this month found no evidence of widespread, economy-wide displacement as yet.
But it did find employment among Americans aged 22 to 25 in AI-exposed occupations was 19 per cent below where it would have been if it had kept pace with less-exposed workers.
The gap appeared to be driven mainly by weaker hiring, although the researchers stressed their findings were descriptive and did not prove AI caused the decline.
The International Labour Organisation estimates one in four jobs worldwide is exposed to generative AI, but says transformation remains more likely than outright replacement.
The International Monetary Fund has estimated almost 40 per cent of global employment could be affected, warning that AI will probably worsen inequality in most scenarios unless governments intervene.
The Luddite case
Fears about machines destroying livelihoods hardly began with ChatGPT, but the acceleration in just four years has propelled the conversation to the foreground.
From your local beer garden to the bus stop, a conversation about AI is never too far from earshot.
Back in the 19th century, the original Luddites were often protesting against employers who used machinery to cut wages, evade apprenticeships and drive down standards.
Their name survived as an insult for anyone standing between an inventor and a profit margin.
In 1930, economist John Maynard Keynes warned of “technological unemployment” caused by labour-saving discoveries arriving faster than society could find new work.
Decades later, computer scientist Joseph Weizenbaum argued that some decisions should never be handed to machines, even if computers became technically capable of making them. Judgment involving compassion and human responsibility, he believed, was different from calculation.
Those concerns were routinely waved away as Luddism or paranoia from technologically illiterate peasants reading too much sci-fi.
History gave a strong counterargument to their fears, in that tractors, factories and computers destroyed particular jobs while helping create industries and occupations their displaced workers could scarcely have imagined.
But Gates’ case, like several others warning of accelerationism, is that AI breaks the pattern both because it’s moving so quickly, and that it could one day potentially substitute our greatest asset: our brains.
“AI for the first time can replace and even exceed human cognition,” he wrote.
The old fears have consequently moved from society’s fringe to our most powerful institutions, with the conversation increasingly gravitating around the rapid concentration of power that theoretically follows a world that sees millions displaced by a handful of companies’ software.
‘Some decisions must remain forever human’
The US now officially describes itself as being in an AI race. The Trump administration’s AI Action Plan says American leadership is necessary to secure economic and military dominance as China develops competing systems.
The US intelligence community’s 2026 threat assessment says AI is already influencing battlefield targeting and military decision-making, while increasing the power of cyber operations, weapons design and unmanned vehicles.
A previous assessment found pro-China actors had used AI-generated newsreaders, profile pictures and fake social-media accounts to aggravate divisions inside the United States.
At the United Nations, Secretary-General António Guterres has warned that artificial intelligence and autonomous weapons are advancing faster than the world can govern them.
“Some decisions must remain forever human – none more than taking a human life,” he said in July.
Gates sees the same arms race unfolding in cyberspace.
“The smartest cybersecurity experts I know are scared about the next few years, because the attackers are getting powerful new capabilities faster than the defenders can fix all the weaknesses,” he wrote.
He has long warned AI could also make it easier to design a deadly disease, manipulate public opinion and launch attacks against hospitals, financial institutions, water systems and power grids.
Then came the most chilling proposition of all.
“As the models become more powerful, they could begin to act against our interests and we could lose control,” Gates wrote.
Thoughts? alexander.blair@news.com.au