The Digital Divide in 2025: How AI is Widening Global Inequality
Nov 12, 2025
Daniela Mata
Some weeks ago I attended a webinar on the International AI Safety Report 2025 hosted by the Mexican Council on Foreign Affairs (COMEXI). It was an interesting talk with many valuable insights into the risks that AI represents, as well as the benefits it generates.
Despite the main theme being risks and safety, one of the concluding remarks and agreements from the report and the webinar itself was that we have agency over AI and people are the ones making the decisions on its development. This idea stuck out to me immediately. Who actually has this agency? It seems like it resides in a select few, and the rest of us are left to figure out how to face the effects of this evolving tool.
Furthermore, the term AI democratization is used often by companies and governments to present a more positive outlook on people's involvement in AI development, profits, and usage. When we reflect further, the reality is not so bright.
The digital divide continues to exist and it's only further extended due to AI. This is because the basic issues of the divide continue to be sidelined in efforts to highlight the issues and sectors that have indeed been addressed and solved thanks to AI. A simple example is how AI has been used to develop voice to text tools to help the deaf and hard of hearing community, yet at the same time it has neglected the people in this community who don't even have stable access to electricity or a smartphone.
According to the Digital 2025 Global Overview Report, 2.63 billion people still lack access to the internet in 2025. These people are the ones who remain most isolated from technological development, and are not only unable to reap the benefits of AI, but they are also negatively affected by the exacerbated inequalities.
Additionally, conscious usage of AI is partially restricted to tools like ChatGPT and DeepSeek while "much of the current AI activity happens "behind the scenes", within broader service offerings" (Kempt, 2025) with products and services integrating AI into their functions. Most people are unaware of how much of the tools they use on a daily basis are powered or influenced by AI, and it is this lack of understanding that leads to growing digital illiteracy.
Therefore, when claims such as "AI makes technology more accessible" are stated, I remain deeply skeptical. Unless the core inequalities that widen the digital divide are addressed, AI on its own cannot lead to higher rates of technological access. Moreover, if we can't ensure an inclusive and equal development, usage, and profits of AI, then the existing inequalities will only continue to grow while giving more power to the elites who, unlike the majority of the population, can have agency over AI.
AI is not here to fix the systemic issues that plague the current society and world order because it wasn't made to do so. Therefore, it is up to us to advocate for our rights, be educated on the new implications AI brings, and support the communities which are left behind due to the digital divide.
-
This is a brief post on the issue, but it's something I'm deeply passionate about and would love to discuss with others. If you have any comments or ideas that you'd like to share about this topic and related ones, please don't hesitate to reach out: