Everyone's excited about AI (except me?)

After 10 years in tech, I’ve been amazed at how quickly AI is transforming the industry, driving rapid innovation across every vertical. It feels like every day I come across a new startup that has built out AI for some sort of specific use case and eagerly added .ai to their domain. I also see less and less startups get any funding if their product does not encompass AI.

When AI first became a thing, I (like many others) rolled my eyes and said - "no, no, it's not going to replace people," "It probably won't be that good." Now, AI has improved at an alarming rate, unlocking all kinds of doors for people across the world. From content writing to code, it's becoming less obvious whether it's written by a person or computer.

While AI's wide range of capabilities are impressive to say the least, there are detrimental impacts not just to tech but to the entire world that I feel aren't being addressed adequately by businesses or governmental entities. AI Models should take social responsibility for the negative impacts they are causing.

Below are my "hot takes" or claims, which I have supported with sources.

?? AI has a Detrimental Impact on the Environment ??

A request made through ChatGPT, an AI-based virtual assistant, consumes 10 times the electricity of a Google Search (Source: International Energy Agency, 2024)

Data Centers are the powerhouse behind AI.

Data centers already require immense amounts of energy, but AI significantly increases this demand. The majority of this energy still comes from fossil fuels, worsening carbon emissions.

Currently, Data centers account for about 1% of the world's total electricity consumption (Source: International Energy Agency, 2024). This demand is growing at an alarming pace.

Beyond Energy - Water ??

Most people are aware of the environmental impact from energy usage, but water usage is often overlooked.

Data Centers must keep electronic components cool, which consumes significant amounts of water. They also use water for off-site generation of electricity.

?? Freshwater scarcity is one of the most pressing global challenges, already affecting 4 billion people annually. Without effective management, severe water stress could impact nearly half the global population by 2030. (Source: Making AI Less “Thirsty,” 2023)

Raw Material Mining ??

The infrastructure of AI is dependent on raw materials including lithium, cobalt, and rare earth materials. These materials are primarily mined in developing regions, often under hazardous conditions that exploit workers, degrade ecosystems, and harm local communities.

Mining operations often cause deforestation, habitat loss, and soil erosion.

There are multiple other considerations including raw material mining, pollution, and the impact on local communities. Many of these data centers are located in areas where the excessive usage of water and electricity pose a significant risk for local populations.

As companies race to compete to strengthen their AI models, it often results in an even larger increase in e-waste.

??? Only about 20% of global e-waste is formally recycled, with the rest incinerated or sent to landfills. (World Health Organization, 2024)

It's imperative we move to more sustainable measures for AI (more addressed in the conclusion).

Social and Economic Impacts

AI Encourages Over-Consumption and Erodes Critical Thinking

Since the introduction of tools like ChatGPT, many of us have developed the habit of turning to AI for instant solutions to challenges at work. While this convenience is undeniably useful, it often comes at the cost of deeper engagement with problems.

Relying too heavily on AI can lead to a decline in our critical thinking and problem-solving skills. The process of brainstorming, collaborating with others, and reflecting on potential solutions is invaluable—it often yields more creative and thoughtful outcomes. However, AI short-circuits this process, presenting a ready-made answer that we may accept without questioning.

This can foster a passive mindset. If AI doesn't have the answer, we might assume the solution doesn't exist or isn't worth pursuing. But many challenges, especially complex or subjective ones, require human intuition, creativity, and nuanced judgment—qualities AI cannot replicate.

AI Takes More Jobs than it Creates

I really hope I'm wrong about this, but so far, it seems like AI is replacing a lot of human-driven work rather than creating new opportunities. While this shift could be a win for business owners- helping them cut costs, streamline operations, and reduce overhead- it raises concerns for workers and, on a larger scale, the tech industry.

Up to 800 million global workers could be displaced by automation and AI by 2030, with about 375 million of them needing to transition to new occupations. (Source: McKinsey and Co)

Many routine tasks, from customer service to data entry, are already being automated by AI, leading to job losses in roles that were once considered stable. And while new industries may emerge, it's hard to ignore that the pace of AI advancement outstrips the pace at which new jobs are being created. If businesses rely too heavily on AI for cost-saving, the question becomes: Are we sacrificing long-term job growth and innovation for short-term efficiency?

Gartner predicts that by 2025, AI will replace nearly 69% of managerial tasks, leading to job displacement in sectors where human decision-making is traditionally seen as essential.

AI is Dangerous

We’ve all seen what AI can do—mimicking human voices and videos for fraud, for instance.

Consider the disturbing trend where scammers replicate someone’s voice to demand money from their loved ones. The accuracy and believability of AI-generated voices make these schemes incredibly effective, posing a danger even to tech-savvy individuals.

We also should be thinking about the data we share with AI platforms? Companies like OpenAI initially operated as nonprofits but have since transitioned into for-profit businesses, raising concerns about how user data is monetized.

Can AI Companies Sell Your Data?

If you’re using AI tools outside of California or Europe, technically, they can sell your data. While about 190 countries have some AI regulations, they often fall short of addressing privacy concerns.

Steps Toward Sustainable AI

What can you do?

Acknowledge the Challenges:

I am not going to tell people to stop using AI, because at this point, it's completely unavoidable. As a small business owner who actively tries to offer clients the most modern and realistic techniques to achieve their business goals, it's not realistic for our team to avoid AI. However, when it comes to searching for information for example, maybe we use Google instead of ChatGPT. When we collaborate as a team to solve problems, let's use our brains first, AI later. While AI is extremely helpful, it oftentimes lacks context, resulting in poorer business decisions.

Support Responsible Companies

When it comes to partnering with tech companies - let's prioritize companies that have a great product but also care about the environment and take social responsibility for how they are building AI.

Push for Better Standards

If you're a Product leader or Engineering leader and you are planning your roadmap for 2025, let's build AI that's sustainable and solves problems. It seems like many AI features these days are built just to say "Yeah, we have AI" instead of because the product we are building is actually solving a business problem.

Protect Yourself:

There are also some ways to protect yourself and your data:

• Avoid unnecessary sharing with AI products.
• Opt-out of data sharing or request data deletion when you can.
• Read the privacy policy when you sign up for a new product leveraging AI
• Share feedback directly with companies/solutions you pay for about data security concerns

Regarding the environmental impacts, the UNEP recommends these measures:

1. Development of standard methods for measuring the environmental impact of AI across the world
2. Tech companies should work to make AI more efficient, reducing energy usage and leveraging recycled water/materials when possible
3. Consumers, businesses, and government entities should pressure data centers to be more green

How has AI impacted your business? Do you consider environmental impacts when planning your product roadmap? Like and Comment - would love to hear from you!

#artificialintelligence #ai #environmentalimpacts #aiandtheenvironment #sustainability #tech #aitech #socialresponsibility #aiimpacts