On 9 November 2026, the day COP31 opened in Antalya, Türkiye, I joined a COP31 virtual panel on Climate Justice, Sustainable Finance and the Future of AI. The other speakers covered climate justice and finance, including Maggie Mullen, Finance Director at re-earth initiative. My job was the AI perspective, and I wanted to give an honest one rather than the usual story where AI arrives as the hero. The company-side note is on the AI Walay blog as AI Walay founder speaks on a COP31 panel.
The one idea I came back to
"AI is both a climate tool and a climate load, and justice decides which one it becomes."
I build AI agents for a living, so I see both sides every week. The same model that books an appointment or answers a customer also runs on electricity and water in a data centre somewhere. It is why I put cheap, deterministic steps in front of model calls, as I describe in AI agents in production. The question I put to the panel was not whether AI is good or bad for the climate. It was who AI is built for, who pays its costs, and how the benefits reach the people on the front line.
The honest footprint
The International Energy Agency estimates that data centres used about 415 TWh of electricity in 2024, around 1.5% of the world total, and projects roughly 945 TWh by 2030 in its base case, just under 3%, with AI the main driver. Any AI tool we build carries part of that footprint, and pretending otherwise would be greenwashing. COP31's presidency has acknowledged this too: the voluntary Antalya AI Pledge, released on 27 October 2026, covers AI's energy and water use.
Where AI genuinely helps
The uses I think justify that footprint are the ones that were impossible a few years ago:
- Flood forecasting that now reaches people in more than 100 countries.
- Balancing power grids that run on variable solar and wind.
- Spotting deforestation and land change from satellite images.
- Giving a farmer advice in her own language, as a voice note on a basic phone.
Why I talk about Pakistan
I come from Pakistan. In 2022, floods put about a third of the country under water and affected around 33 million people, with damage and losses of roughly USD 30 billion. In 2025 the monsoon floods killed more than 1,000 people. Pakistan produces less than 1% of global greenhouse gas emissions. That is what climate injustice looks like, and it is also where well-built AI could save the most lives, if it is built for those communities and not only for the richest markets.
Three justice gaps
The gaps that decide whether AI helps the front line:
| Gap | What it means | What would close it |
|---|---|---|
| Data | The most exposed regions have the fewest weather stations and the fewest models trained on their reality | Local data collection and open datasets |
| Language | Most AI works best in English; a warning only helps if people understand it | Voice-first tools in Urdu, Sindhi, Swahili, Spanish and more |
| Ownership | Compute and models sit with a few countries and companies | Community-owned data and open tools |
This is where my own work connects. I build WhatsApp and voice agents, and those are the channels people in the Global South actually use. The technology that books appointments for a business can deliver an early warning or a crop advisory. The hard part is not the model. It is designing for low bandwidth, local language and trust.
AI and climate finance: making the money traceable
At COP29, countries agreed a new goal of USD 300 billion a year by 2035 for developing countries, plus a roadmap towards USD 1.3 trillion a year from all sources. Money at that scale only flows if investors and communities trust the numbers. AI can help with that: checking restoration and forest-carbon projects from satellite data, reading thousands of sustainability reports to flag claims that do not match the evidence, lowering the cost of climate disclosure for smaller firms, and tracking whether adaptation money reaches the communities it was promised to.
How leaders and negotiations stay accountable
One of the questions was how to keep leaders accountable when major emitters step back. My answer was transparency that does not depend on any one government: independent, AI-assisted monitoring of emissions, finance flows and pledges, published openly so that progress, or the lack of it, is visible to everyone. Negotiations work better when they start from shared, verifiable data rather than competing claims.
My advice to young people
You do not need permission to build. Pick one real problem in your own community, learn the tools and ship something small that helps. A WhatsApp bot that sends a flood alert in the local language is worth more than a perfect pitch deck.
Closing
AI will not solve the climate crisis by itself, and it should not become another source of emissions that the poorest countries pay for. But if we build it for the front line, in local languages, with honest accounting of its footprint and with money that reaches the people it was promised to, it can be one of the most powerful tools we have. Thank you to the organisers and my fellow panellists for bringing technology, justice and finance into one conversation.
If you are working on a climate, early-warning or finance-transparency tool and want an AI build that works on basic phones and in local languages, book a call below.
Key takeaways
- AI is both a climate tool and a climate load; justice decides which one it becomes.
- The IEA estimates data centres used about 1.5% of world electricity in 2024, rising to just under 3% by 2030.
- Pakistan emits under 1% of global greenhouse gases, yet the 2022 floods affected about 33 million people.
- The data, language and ownership gaps decide whether AI reaches the front line.
- AI-based measurement can make climate finance traceable, which builds the trust large flows need.
Frequently asked questions
- What was the COP31 panel about?
- It was a COP31 virtual panel on Climate Justice, Sustainable Finance and the Future of AI, held online on 9 November 2026, the opening day of COP31 in Antalya, Türkiye. Speakers covered climate justice, sustainable finance and AI, and I gave the AI perspective as founder of AI Walay.
- How much electricity does AI use?
- The International Energy Agency estimates that data centres used about 415 TWh in 2024, around 1.5% of global electricity, and projects about 945 TWh by 2030 in its base case, just under 3%. AI is the main driver of that growth, which is why honest footprint accounting matters.
- How can AI support climate justice?
- AI supports climate justice when it closes the data, language and ownership gaps: models trained on data from the most exposed regions, warnings and advice delivered by voice in local languages on basic phones, and communities that own their data and can use open tools.
- How can AI help climate finance?
- AI can verify projects from satellite data, flag greenwashing in sustainability reports, cut the cost of climate disclosure for small firms and track whether adaptation and loss-and-damage money reaches the communities it was pledged to. That traceability builds the trust that large finance flows depend on.
- What advice did you give young people?
- You do not need permission to build. Pick one real problem in your own community, learn the tools and ship something small that helps. A WhatsApp bot that sends a flood alert in the local language is worth more than a perfect pitch deck.
Sources
Written by Syed Husnain Haider Bukhari
AI engineer, data scientist, and founder of Revolutionary Technologies LLC. Ships production AI agents, automations, and data platforms for teams in the US, UK, and UAE — including AgentFlow, AI Walay, and ProLeads.
Get in touch →Related pages
Want this built instead of researched?
I build these systems for teams in the US, UK, and UAE. Book a free 30-minute consultation and you get a one-page plan and a fixed-scope quote within 48 hours — or message me directly, whichever is faster for you.
Prefer a form? Send a project brief →