Bill Gates proposed taxing robots in a 2017 Quartz interview, arguing that automation displacing workers should face levies equal to the income taxes those workers paid. The WEF projects 92 million jobs lost globally and 170 million created by automation. The gap between lost and created is where displaced workers fall through.
This is a growing concern, and one we are not talking about enough.
Bill Gates thinks we should tax the robots. In a 2017 interview with Quartz, he made the argument simply: if a human worker earning $50,000 a year is taxed on that income, why shouldn’t a robot doing the same work face a similar levy? Take the revenue from that tax, and use it to fund retraining programs, social services, and eventually, something like Universal Basic Income for the workers left behind.
It is a reasonable starting point. And I think it points in the right direction.
But it is not that simple.
AI replacing jobs is already happening — and the numbers are not reassuring
AI replacing jobs is not a future scenario. It is already underway. A 2023 survey conducted across 31 countries found that more than half of all respondents said they felt nervous about AI’s impact on their employment. That is not a fringe concern. That is a majority.
The World Economic Forum’s Future of Jobs Report 2025 offers a headline that sounds positive: 92 million roles will be lost to automation globally, but 170 million new ones will be created, for a net gain of 78 million jobs. Read that twice and it sounds fine. But the report describes what is actually happening on the ground. Over 95,000 tech workers lost their jobs in 2024 alone. The roles being displaced first are not at the top of the income ladder. Customer service agents, data entry clerks, junior analysts, call center workers — the positions that sustain middle-income households in economies like ours — are the ones going first. The new roles being created tend to require skills that displaced workers do not currently have and cannot acquire quickly without deliberate, funded support.
That is the transition gap. It is not the net number that keeps me up at night. It is the gap between when the old jobs disappear and when the people who held them land somewhere new.
The questions the robot tax debate has not answered
The robot tax proposal sounds clean in theory. In practice it generates four questions that nobody in the policy conversation wants to sit with.
What exactly counts as a “robot”?
A physical machine on a factory floor is easy to identify. But the AI replacing jobs in the Philippines right now is mostly software. It is the voice AI that handles the first three minutes of a customer support call. It is the tool that now drafts the legal brief before a junior associate touches it. It is the algorithm that flags claims before a human adjuster reviews them. When the “robot” is a cloud subscription that a company pays $30 a month for, how do you define it, tax it, or even track it?
Does taxing automation slow it down — and is that good or bad?
Gates himself suggested that a robot tax could slow automation “just a bit,” giving governments and communities more time to adapt. Some economists see that as the feature. Others see it as penalizing productivity and competitive advantage. Neither side is obviously wrong. The honest answer is: it depends on how it is designed, and right now nobody has designed it well enough to know.
What stops companies from moving the work offshore?
This is the arbitrage question. If the Philippines implements a robot tax and Vietnam does not, companies route the automated work through Vietnam. This is not speculation. It is what happened for decades with labor cost arbitrage, and there is no reason to think automation arbitrage would behave differently. A robot tax without international coordination does not solve the problem. It just shifts it across borders while leaving domestic businesses at a disadvantage.
If we implement Universal Basic Income, how do we make sure it actually reaches the workers who need it?
Gates raised the possibility of UBI in a 2017 Reddit AMA, noting it might become viable as the US becomes wealthier. Others have linked his robot tax proposal to UBI as a natural funding mechanism. The idea has backing from unlikely allies: Elon Musk, Mark Zuckerberg, and various economists have all raised versions of it. But UBI is not a new idea, and its weakest link has always been the same one: governance. Money allocated to social programs has a history of arriving somewhere other than where it was intended. The question of how to design accountability into a UBI system is not a small technical detail. It is the whole problem.
What business leaders in the Philippines should actually do right now
Waiting for governments to resolve the robot tax debate is not a strategy. The policy conversation moves slowly, and the displacement is happening now. Here is what I would actually recommend to the leaders and teams I work with.
Map your team’s exposure before the pressure arrives. The first step is not restructuring. It is clarity. Which roles in your organization are most likely to be affected by AI replacing jobs over the next two to three years? Not so you can start cutting, but so you can start planning. In our work at PAIBA helping organizations across the Philippines with AI adoption, the teams that navigate workforce transitions well are consistently the ones that saw the exposure coming. The ones that got caught off guard either moved too fast and damaged morale, or moved too slow and lost their competitive position.
Invest in transition before it is urgent. Reskilling under duress is expensive and ineffective. People are in survival mode, retention drops, and the organizational knowledge you are trying to preserve walks out the door. At Olern, the learning platform we have built for the Philippine market, one pattern shows up consistently: reskilling works when it is positioned as growth, not as a response to a threat. Build the capability now, while there is still runway. Ask your HR and L&D leads: what does our team need to be able to do in two years that they cannot do today?
Start the policy conversation at the industry level, not the national one. Generic advocacy for “AI governance” gives policymakers room to stay vague. The BPO sector, for example, can be specific: here is the number of roles at risk, here is what those workers need, here is what retraining costs, here is what the timeline looks like. That kind of specificity is harder to dismiss in a committee. It also positions your industry as a responsible actor rather than one waiting to be regulated.
Design transition support into your AI deployments, not after them. When a deployment replaces a workflow or a role, the transition plan for the people affected should be part of the deployment plan — not a conversation you have six months later. This is where the robot tax conversation, at the company level, can actually happen. What percentage of the productivity gain from this deployment goes toward the people it displaces?
The bigger issue
One thing is clear: AI replacing jobs is not slowing down.
But the bigger issue is not the robots themselves. It is whether we are doing enough to prepare the people who will be most affected — before the transition gap turns into a freefall. That requires governments who can design accountable programs. Companies willing to invest in their workers before the layoff becomes necessary. And a public conversation that gets more specific than “tax the robots and hope for the best.”
Are we having that conversation? Not yet, really. The window to have it while the numbers are still manageable is closing.
What do you think?
If you are working through AI adoption or workforce planning in your organization, I would be curious to hear what you are seeing. And if this raised questions you want to work through, reach out — this is exactly the kind of conversation we are built for at PAIBA.
Frequently Asked Questions
What did Bill Gates actually propose about taxing robots?
In a 2017 interview with Quartz, Bill Gates proposed that companies using robots or automation to replace workers should pay a tax equivalent to what those workers paid in income taxes. Gates argued that this tax could slow the pace of automation slightly and fund social programs and retraining for displaced workers. He raised the idea of Universal Basic Income separately, in a 2017 Reddit AMA, noting it might become viable as societies become wealthier.
Is AI replacing jobs a real risk in the Philippines?
Yes. The Philippines is particularly exposed because of its large BPO sector, which employs hundreds of thousands of workers in customer service, data processing, and back-office roles — exactly the categories most affected by AI job displacement. A 2023 survey across 31 countries found over half of respondents felt nervous about AI’s impact on their jobs, and the WEF Future of Jobs Report 2025 projects that 92 million roles will be displaced globally before new ones are created.
What is the transition gap in AI job displacement?
The transition gap is the period between when AI replacing jobs eliminates existing roles and when displaced workers find new employment. The World Economic Forum projects a net gain of 78 million jobs from automation globally, but the workers losing jobs and those filling new ones are rarely the same people. Without reskilling investment and policy support, the transition gap can mean extended unemployment for workers in middle-income roles.
What are the main arguments against a robot tax?
Critics raise three main objections. First, defining what qualifies as a taxable “robot” is difficult when displacement comes from software rather than physical machines. Second, a robot tax could slow productivity growth and put businesses in countries with the tax at a disadvantage relative to competitors in countries without it. Third, without international coordination, companies can route automated work through lower-regulation jurisdictions, reducing the tax’s effectiveness.
How can Philippine businesses prepare their teams for AI job displacement?
Philippine business leaders can take four concrete steps: audit which roles are most at risk of AI displacement in the next two to three years; invest in reskilling programs before workforce disruption is urgent; engage industry associations to develop specific policy proposals rather than generic calls for governance; and build worker transition plans into AI deployment projects from the start, not as an afterthought. Organizations like PAIBA work with Philippine companies on exactly this kind of workforce readiness planning.
Is Universal Basic Income a realistic solution to AI job displacement?
UBI has growing support from technologists and economists as a response to automation-driven job loss, but its viability depends on two unsolved problems: sustainable funding mechanisms and accountable distribution systems. The robot tax proposal is often linked to UBI as a funding source, though Gates originally proposed them separately. Pilot UBI programs in various countries have shown mixed results, and designing a system that delivers benefits to displaced workers without significant leakage to inefficiency or corruption remains the central challenge.
