The Gates Foundation announced this month that it will spend at least $1 billion over the next two years on AI projects in healthcare, education, agriculture and digital infrastructure. In its official announcement, the foundation said 40% of the funding will go to education, another 40% to healthcare, 10% to agriculture and 10% to the underlying digital infrastructure needed to make AI accessible more broadly.
At the same time, Gates has become increasingly outspoken about the risks of moving too fast.
Reuters reported that he believes governments around the world are “way behind” in preparing for AI’s impact, including job displacement, cybersecurity threats and other potentially severe social consequences.
Gates Is Not Arguing Against AI
The tension becomes easier to understand when separating two different questions: how powerful AI could become and where that power should be deployed.
Gates has not abandoned the view that AI can improve productivity and living standards. His foundation is specifically targeting areas where access remains limited, including clinical decision support, personalized education and agricultural advice.
The new funding plan is substantial. The foundation says its AI initiatives are intended to help tools work across more languages, adapt them to local conditions and make them available to communities that would otherwise be left behind.
A separate coalition announced this week includes 60 organizations working toward giving roughly 3.4 billion people access to AI tools in languages currently underrepresented by major models.
| Area | Share |
|---|---|
| Education | 40% |
| Healthcare | 40% |
| Agriculture | 10% |
| Digital infrastructure | 10% |
The Risk Argument Is About Capability, Not Usefulness
Gates' warning is therefore not that AI has no value.
It is that a technology capable of producing major benefits could also become dangerous if deployed faster than governments can build rules around it.
That concern is no longer limited to Gates. A recent Reuters/Ipsos poll found 73% of Americans worry AI companies are not doing enough to prevent potentially catastrophic outcomes, while 55% support slowing AI development.
The industry itself is also becoming more divided. Reuters reported that executives and researchers at major AI labs have recently pushed for stronger safeguards or slower frontier development amid concerns around model autonomy, cyber misuse and loss of human oversight.
Coinpaper has already examined how AI spending is becoming increasingly capital-intensive through data-center debt and why the sector's financial risks are expanding alongside its technical capabilities.
